《当古代圣贤遇上现代牛马》

subject_definitions:
<Subject 1> is the first female student from <Picture 1>, preserving her youthful adult face, low braided dark updo, ink-blue bronze-floral robe and slim natural build; her current standing posture and empty hands come from <Picture 4>.
<Subject 2> is the male teacher from <Picture 2>, preserving his adult face, long dark hair, sparse mustache and chin beard, dark robe, white inner collar and large wooden beads; the open book carried in his left hand comes from <Picture 3> and <Picture 5>.
<Subject 3> is the same classroom and seated classmates from <Picture 3>, with low floor desks, woven cushions, a single six-woman row and warm directional window light.
<Picture 3> is the exact opening composition of [Shot 1] at 0.00 seconds.
<Picture 4> is the composition and standing-state anchor for [Shot 2]; it is a planned cut, not a frame-interpolation destination.
<Picture 5> is the composition and interaction-state anchor for [Shot 3], with both characters standing beside the first student's place.

summary:
[keyframe completion + reference generation] Create one 15.00-second, three-shot classroom comedy round. <Subject 2> asks a proverb question, <Subject 1> rises and gives a confidently absurd answer, and their contrasting reactions share a clear two-person frame. <Picture 3>, <Picture 4> and <Picture 5> define the chronological shot anchors; <Picture 1> and <Picture 2> supply identities, not opening portrait shots.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - retain identity, robe, hairstyle and original seat location; standing develops visibly from sitting.
<Subject 2> (appears in [Shot 1], [Shot 3]): fully_preserved - retain identity, clothes, beads and left-hand open book; his approach is an explicit action.
<Subject 3> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - retain physical classroom, window source, floor seating and classmates' positions.
<Picture 3> ([Shot 1] opening frame): fully_preserved - match the user's master composition before movement begins.
<Picture 4> ([Shot 2] composition anchor): fully_preserved - retain standing woman, lower seated neighbor and screen-right window light.
<Picture 5> ([Shot 3] composition anchor): fully_preserved - retain both clear faces, teacher screen-right, student screen-left, open book and teaching gesture.

detailed_description:
Maintain refined photographic realism, muted ink-blue clothing, teal-gray plaster and deep brown timber, with local warm window highlights and natural peach skin. Preserve both identities, plausible adult scale, the low desks and classroom geography across cuts; no subtitles, narration, fantasy effects or generated freeze frames.

[Shot 1] Begin from <Picture 3>: the classroom master shows <Subject 2> seated centrally on the low dais, holding the open book in his left hand, and <Subject 1> seated at the far left of the six-woman row. From 0.40 to 2.20 seconds, <Subject 2> (S1) asks clearly, <d>[Chinese] 富贵什么求？</d> Use exactly this fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. After the question, he lowers his right hand, rises naturally using the desk for balance, keeps the book open in his left hand, steps down and walks around the desk toward the first woman's position. At the same time, she places her brush on her own low desk and rises behind it. Establish both movements visibly in this wide shot; nobody changes places instantaneously. The other five women remain seated and glance toward the first speaker. Quiet fabric and floor sounds accompany the movement.

[Shot 2] At 00:06.800, cut to the standing-answer composition of <Picture 4>. <Subject 1> is upright, hands loosely together at her belt, looking slightly screen-right toward the approaching teacher. Her seated pale-sage neighbor remains lower at screen-left and looks up at her. From 7.00 to 8.30 seconds, <Subject 1> (S2) answers, <d>[Chinese] 双色球！</d> Use exactly this fixed voice: A young adult female light mezzo voice in a mid-high register, clear warm forward resonance, almost no breathiness, light precise consonants and neat sentence endings. Use a measured, confidently matter-of-fact pace with a mild natural Northeastern Mandarin cadence. Keep this vocal age, register and timbre stable; no childish squeak, coy softness or exaggerated comedy voice. Let all syllables finish cleanly. From 8.30 to 9.50 seconds, her lips close and her chin lifts slightly in calm certainty; preserve a clear, nearly still reaction for the editor's freeze beat. Her neighbor gives a small approving nod. The teacher finishes his already-established approach outside this tighter frame.

[Shot 3] At 00:09.500, cut to <Picture 5>, the standing two-person close shot. <Subject 1> remains screen-left at her original place; <Subject 2> stands beside her at screen-right, slightly inclined toward her. Both three-quarter faces stay sharp. He keeps the same open book in his left hand and raises his right index finger above it, beginning a puzzled correction with visible mouth movement but no additional intelligible dialogue. She listens seriously, then gives one small confident nod. Her agreement makes him pause mid-gesture, eyebrows tightening, lips stopping slightly apart. From 13.20 to 15.00 seconds, hold their readable contrasting expressions with only natural breathing. Keep both standing, his book open in the left hand and her hands together. End before she sits or the second woman rises.

overall_soundscape:
Continuous quiet indoor room tone, faint outdoor birds, soft robe movement and restrained footsteps. Only the two specified lines are intelligible speech; the final correction uses mouth movement and a small nonverbal exhale. No narrator, audience laughter or additional spoken answers.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the second young adult female student: face and low braided hairstyle from <Picture 1>, pale-sage classroom robe, cream collar, pale ribbon and empty-handed standing posture from <Picture 5>. Do not copy the dark-green garment or seated portrait pose of <Picture 1>.
<Subject 2> is the teacher from <Picture 2>, retaining his long dark hair, sparse facial hair, dark robe, white collar and wooden beads. The same open book is supported by his left hand in <Picture 4> and <Picture 5>.
<Subject 3> is the first ink-blue-clad female student from <Picture 4>; she sits at her existing place and does not answer in this round.
<Subject 4> is the classroom from <Picture 3>, with the fixed six-student row, floor cushions, low desks and physical window source; <Picture 4> and <Picture 5> supply the reverse-view light and material continuity.
<Picture 4> is the exact composition anchor at 0.00 seconds of [Shot 1], before the first student sits and the teacher changes places.
<Picture 5> is the standing two-person composition anchor for [Shot 2] at 5.60 seconds, reached after the visible movement and rising.

summary:
[keyframe completion + reference generation] Create one 12.00-second, two-shot classroom comedy round. <Subject 3> sits, <Subject 2> moves screen-right to screen-left toward the next student while asking, and <Subject 1> rises before answering earnestly. The final two-person reaction contrasts her sincerity with his restrained recoil. <Picture 4> starts the video; <Picture 5> anchors the later dialogue shot. <Picture 3> supplies geography only, not a new opening wide shot.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - retain her distinct face, low braided hair, pale ribbon and classroom costume; show sitting develop into standing.
<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - retain identity, clothes, beads and left-hand open book, with an explicit move to the adjacent student.
<Subject 3> (appears in [Shot 1]): fully_preserved - retain her identity and ink-blue robe while she sits silently at the first place.
<Subject 4> (appears in [Shot 1], [Shot 2]): fully_preserved - preserve the room, seating order, low desks and physical light source.
<Picture 4> ([Shot 1] opening composition): fully_preserved - establish its standing pair before their subsequent movement.
<Picture 5> ([Shot 2] composition anchor): fully_preserved - retain the second student screen-left, teacher screen-right, visible faces, robe colors and open book.

detailed_description:
Preserve the references' refined photographic realism, teal-gray walls, deep timber and local warm window highlights. Keep natural adult proportions, readable faces and the same physical window at screen-right in this reverse view.

[Shot 1] Begin exactly from <Picture 4>: <Subject 3>, the ink-blue first student, stands screen-left beside <Subject 2>, the dark-robed teacher screen-right. His left hand supports the open book. Over the first two seconds, she sits onto her original cushion while he lowers his right hand, turns toward the next desk and takes two ordinary steps screen-left along the row's front edge. The camera widens and tracks left, revealing <Subject 1>, the second student from <Picture 1>, still seated at her adjacent low desk in the pale-sage robe of <Picture 5>. Preserve <Subject 4>'s seating geography from <Picture 3>; never return the teacher to the dais or morph one woman into another.

From 1.20 to 3.00 seconds, <Subject 2> (S1) asks while approaching her, <d>[Chinese] 夜长什么多？</d> Use exactly this fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. The first student remains silent, settles her robe and follows him with her eyes. Other visible classmates stay seated and make small attentive glances. From 3.00 to 5.40 seconds, the second student places her brush on her own low desk and rises naturally in place. He waits beside her with the book still open in his left hand. Show her rising before cutting; no instantaneous standing or exchange of faces.

[Shot 2] At 00:05.600, cut to <Picture 5>'s two-person composition: <Subject 1> stands screen-left, empty hands gently joined near her waist, and <Subject 2> stands screen-right. Both three-quarter faces remain clear. From 5.80 to 6.80 seconds, <Subject 1> (S2) answers, <d>[Chinese] 尿多！</d> Use exactly this fixed voice: A young adult female upper-register voice with soft rounded resonance and a small natural airy edge, gentle consonant attacks and plainly grounded endings. Speak earnestly at an easy pace with a mild natural Northeastern Mandarin cadence. Preserve this airy rounded timbre and adult vocal age; no baby voice, squeal or flirtatious delivery. Finish both syllables cleanly. From 6.80 to 7.60 seconds, she closes her mouth and holds an entirely sincere, matter-of-fact expression for the editor's answer freeze beat.

From 7.60 to 10.00 seconds, the teacher retracts his open right hand, leans back slightly and presses his lips together; his eyes widen as he takes a small breath to object. His mouth starts a brief wordless correction, then closes as if he has swallowed it. The woman remains composed, eyebrows lifting slightly in innocent puzzlement. Avoid repeating the previous round's raised-index-finger lecture. Keep the left-hand book open and steady; no tea cup, spitting or extra dialogue. From 10.00 to 12.00 seconds, hold both readable contrasting expressions with natural breathing, his right hand relaxed near the book and her hands together. Both remain standing for the following change of student. No generated freeze frames, subtitles or narration.

overall_soundscape:
Continuous quiet classroom room tone, faint outdoor birds, gentle robe rustles, cushion contact and two restrained footsteps. Only the specified question and answer are intelligible speech; the teacher's later objection is silent mouth movement and a small nonverbal breath. No narrator, audience laughter or additional answers.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the third young adult female student: face and center-parted low chignon from <Picture 1>, slate-lavender classroom robe, white collar and empty-handed standing posture from <Picture 5>.
<Subject 2> is the teacher from <Picture 2>, preserving his long dark hair, sparse mustache and chin beard, dark robe, white collar and large wooden beads. His left hand supports the same open book seen in <Picture 4> and <Picture 5>.
<Subject 3> is the second pale-sage-clad female student from <Picture 4>, with her low braided hair and pale ribbon; she sits at her existing seat and remains silent.
<Subject 4> is the classroom and fixed seating arrangement from <Picture 3>, retaining low desks, cushions, dark timber and the physical window; the reverse-view light follows <Picture 4> and <Picture 5>.
<Picture 4> is an edited opening-composition reference for [Shot 1]: retain its standing pair and left-hand open book, but start with the teacher's right hand lowered near the book after the previous round's reaction. It is not an actual video tail frame.
<Picture 5> is the standing dialogue composition anchor for [Shot 2] at 6.00 seconds, after the second student sits, the teacher moves and the third student rises.

summary:
[keyframe completion + reference generation] Generate one 13.00-second, two-shot classroom comedy round. <Subject 3> sits, <Subject 2> moves screen-right to screen-left toward the next seat while asking, and <Subject 1> rises before answering earnestly. Their shared reaction ends with the teacher searching his book for a reply and falling silent. <Picture 4> guides the edited opening composition, <Picture 5> anchors the later dialogue, and <Picture 3> supplies room geography only.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - retain her distinct face, center-parted low chignon, slate-lavender costume and own seat.
<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - retain identity, clothes, beads and left-hand open book; movement to the next student is visible.
<Subject 3> (appears in [Shot 1]): fully_preserved - preserve the second student's identity and pale-sage clothing as she sits silently.
<Subject 4> (appears in [Shot 1], [Shot 2]): fully_preserved - maintain room, seat order and physical light source.
<Picture 4> ([Shot 1] opening composition): partially_preserved - preserve the two standing characters and framing, with the teacher's right hand lowered to match the planned preceding ending.
<Picture 5> ([Shot 2] composition anchor): fully_preserved - preserve third student screen-left, teacher screen-right, clear faces and the same open book.

detailed_description:
Preserve refined photographic realism, muted teal-gray walls, deep brown timber, slate-blue clothing and local warm window highlights. Maintain natural adult proportions, clear faces and the same screen-right physical window in this reverse view.

[Shot 1] Use <Picture 4>'s composition: <Subject 3>, the pale-sage second student, stands screen-left beside <Subject 2>, the dark-robed teacher screen-right. His left hand supports the open book; his right hand is already lowered near it, continuing the previous reaction rather than restarting its raised gesture. Over the first two seconds, she sits onto her original cushion while he turns toward the third seat and takes two steps screen-left along the row's front edge. The camera widens and tracks left, revealing <Subject 1> seated at the adjacent desk, with her center-parted low chignon and slate-lavender robe. Keep <Subject 4>'s seating arrangement from <Picture 3>. Nobody swaps faces or returns to the dais.

From 1.20 to 3.40 seconds, <Subject 2> (S1) asks, <d>[Chinese] 万丈高楼什么起？</d> Use this fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. The second student settles her sleeves and turns her eyes toward the third; other visible classmates remain seated and attentive. From 3.40 to 5.80 seconds, the third student puts her brush on her low desk and rises in place. The teacher waits beside her, the book still supported by his left hand.

[Shot 2] At 00:06.000, cut to <Picture 5>'s standing two-person composition: <Subject 1> screen-left with empty hands joined near her waist; <Subject 2> screen-right with his right fingertips resting by the open page. Keep both three-quarter faces sharp. From 6.20 to 7.50 seconds, <Subject 1> (S2) answers, <d>[Chinese] 买不起！</d> Use this fixed voice: A young adult female low-register voice with cool clean resonance, minimal breathiness, precise separated syllables and slightly deliberate pacing. Use a mild natural Northeastern Mandarin cadence and a level, dryly resigned finish. Keep the voice firm and youthful; no husky texture, sobbing or whispering. She speaks plainly, without joking or crying, then closes her lips. From 7.50 to 8.30 seconds, retain her calm, resigned expression for the editor's answer freeze beat.

From 8.30 to 11.20 seconds, the teacher starts to lift his right fingertips to turn a page, then stops before turning it. He looks at her, glances down at the book as if searching for a reply, and opens his mouth slightly without producing words. Finding nothing to say, he gently presses the same page flat and looks back at her. Keep his face readable when he lowers his eyes; no deep bow, pointing, widened-eye recoil or shouting. She gives a tiny matter-of-fact eyebrow lift, still convinced her answer is correct. From 11.20 to 13.00 seconds, hold their restrained contrasting expressions with natural breathing. Both remain standing; his book stays open in the left hand, right hand on its edge, and her hands stay empty. No extra dialogue, tea cup, subtitles, narration or generated freeze frames.

overall_soundscape:
Quiet continuous classroom room tone, faint distant birds, soft robe rustles, cushion contact and restrained footsteps. Only the specified question and answer are intelligible speech; the teacher's later attempt to reply is silent mouth movement with a faint breath and soft paper contact. No narrator, audience laughter or additional answers.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the fourth young adult woman, with defined cheekbones, long almond eyes, firm brows, a fully coiled high bun with a straight simple pin, a charcoal-olive botanical robe and white collar. Preserve her exact distinct identity from <Picture 1>, with classroom clothing, hair placement and expression from <Picture 5>.
<Subject 2> is the teacher from <Picture 2>, the original sharp identity reference: same adult face, clear eyes, sparse individually resolved mustache and chin beard, long dark hair, black robe, white collar and wooden beads. Preserve his LEFT-hand support of the same open tan book from <Picture 4> and <Picture 5>.
<Subject 3> is the third young adult woman with a center-parted low chignon and slate-lavender floral robe from <Picture 4>; she sits silently at her existing desk.
<Subject 4> is the classroom and exactly six female students' seating from <Picture 3>. Opening-camera left to right: ink-blue low bun; pale-sage low braids and ribbon; slate-lavender low chignon; charcoal-olive high bun; plum-brown long braid; peacock-teal half-up long hair. The reverse dialogue camera sees this order reversed. Preserve that geometry, low desks, cushions, timber and physical windows; do not clone or add pupils.
<Picture 4> is an edited opening-composition reference for [Shot 1], not an actual generated video tail frame. Preserve the pair and left-hand open book, adjusting the right-hand opening pose to the preceding planned ending described below.
<Picture 5> anchors the current woman's standing two-person composition in [Shot 2] and any later return to it, after the preceding student sits, the teacher moves and the current student rises.

summary:
[keyframe completion + reference generation] Generate one 15.00-second classroom comedy round with 4 shots. The previous student sits, the teacher moves along the desk front from screen-right to screen-left and asks, then the current woman rises and earnestly gives her answer. Maintain two readable faces in the answer and shared reaction; the student's certainty and teacher's contrasting response carry the comedy. Identity references are not playback frames, and the room reference does not restart the opening.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4]): fully_preserved - same face, youthful adult age, classroom robe, hairstyle and fixed seat, with the specified restrained performance.
<Subject 2> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4]): fully_preserved - sharp original facial identity, clothes, beads and same left-hand open book; only expression and right-hand gestures change.
<Subject 3> (appears in [Shot 1] and any wider class view): fully_preserved - preserve previous student's appearance and own seat while sitting and reacting.
<Subject 4> (appears throughout): fully_preserved - same classroom, six existing female students, seat order, physical light source and wardrobe relationships.
<Picture 4> ([Shot 1] opening composition): partially_preserved - retain framing and people, with the teacher's hand and shoulders adjusted to the planned preceding ending rather than replaying its questioning pose.
<Picture 5> (standing dialogue composition): fully_preserved - preserve girl screen-left, teacher screen-right, both clear faces, realistic adult scale and book support.

detailed_description:
Preserve photographic realism, sharp equally focused faces, muted teal-gray walls, dark timber, natural warm skin and the same screen-right window in this reverse view.

[Shot 1] Begin with <Picture 4>'s standing composition: <Subject 3> screen-left, <Subject 2> screen-right, his left hand supporting the open book and right fingertips resting on its page edge. She sits onto her own cushion while he turns and crosses the central aisle screen-right to screen-left toward <Subject 1>. Track gently left and widen, keeping <Subject 4>'s desks and seat order unchanged. Nobody returns to the dais or walks through furniture.

From 1.00 to 4.30 seconds, <Subject 2> (S1) asks, <d>[Chinese] 三百六十行，行行出什么？</d> Fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. From 4.30 to 6.20 seconds, <Subject 1> puts down her brush and rises at her own desk. She stands upright, chin slightly lifted, empty hands near her waist. The teacher stops beside her and opens his right palm near his waist. Seated classmates follow him with small eye movements.

[Shot 2] At 00:06.500, cut to <Picture 5>'s clear standing pair, fourth student screen-left and teacher screen-right. From 6.70 to 7.70 seconds, <Subject 1> (S2) answers, <d>[Chinese] 牛马！</d> Fixed voice: A young adult female middle-register voice with bright full resonance, firm clean consonant attacks and brisk decisive timing. Use a mild natural Northeastern Mandarin cadence, projecting an earnest answer without shouting or clipping. Preserve the full midrange timbre; no soprano squeal, breathy softness or masculine growl. She delivers it as a proud, correct declaration, not a prank. From 7.70 to 8.50 seconds, keep her lips closed and posture firm for the editor's answer beat. The teacher looks at her; his open right palm stops moving.

[Shot 3] At 00:09.000, cut wider from the SAME camera side, revealing the existing row. Exactly five other women remain seated at their original desks; all stop writing or adjusting sleeves and turn toward the fourth student together. Nearby classmates give small earnest nods. Preserve their distinct costumes and hair from <Subject 4>; no clones, extra pupils or laughter at the teacher.

[Shot 4] At 00:10.500, return to the same standing pair. Through 13.00 seconds, the teacher scans the class with his eyes and a small head turn, then looks back at her. His right palm rotates slightly in disbelief; his mouth opens to object but no words emerge. He slowly returns that hand to the lower book edge. She remains convinced, giving one tiny confirming nod. From 13.00 to 15.00 seconds, retain both contrasting faces with natural breathing, her hands empty, his book still open in the left hand. No ruler breaking, cup, shouting, subtitles, narrator, extra dialogue or generated freeze frames.

overall_soundscape:
Continuous quiet classroom room tone, restrained footsteps, cloth rustles and cushion contact. The five classmates briefly stop their writing and sleeve sounds together after the answer; then one soft approving breath is heard. Only the specified question and answer are intelligible speech; no laugh track, narration or audible scolding.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the fifth young adult woman, with relaxed almond eyes, a softly rounded oval face, a small mole on her right cheek, loosely gathered hair in one long braid down her back, and a plum-brown floral robe with teal embroidered trim. Preserve her exact distinct identity from <Picture 1>, with classroom clothing, hair placement and expression from <Picture 5>.
<Subject 2> is the teacher from <Picture 2>, the original sharp identity reference: same adult face, clear eyes, sparse individually resolved mustache and chin beard, long dark hair, black robe, white collar and wooden beads. Preserve his LEFT-hand support of the same open tan book from <Picture 4> and <Picture 5>.
<Subject 3> is the fourth young adult woman with a fully coiled high bun, simple straight pin and charcoal-olive botanical robe from <Picture 4>; she sits silently at her existing desk.
<Subject 4> is the classroom and exactly six female students' seating from <Picture 3>. Opening-camera left to right: ink-blue low bun; pale-sage low braids and ribbon; slate-lavender low chignon; charcoal-olive high bun; plum-brown long braid; peacock-teal half-up long hair. The reverse dialogue camera sees this order reversed. Preserve that geometry, low desks, cushions, timber and physical windows; do not clone or add pupils.
<Picture 4> is an edited opening-composition reference for [Shot 1], not an actual generated video tail frame. Preserve the pair and left-hand open book, adjusting the right-hand opening pose to the preceding planned ending described below.
<Picture 5> anchors the current woman's standing two-person composition in [Shot 2] and any later return to it, after the preceding student sits, the teacher moves and the current student rises.

summary:
[keyframe completion + reference generation] Generate one 13.00-second classroom comedy round with 2 shots. The previous student sits, the teacher moves along the desk front from screen-right to screen-left and asks, then the current woman rises and earnestly gives her answer. Maintain two readable faces in the answer and shared reaction; the student's certainty and teacher's contrasting response carry the comedy. Identity references are not playback frames, and the room reference does not restart the opening.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - same face, youthful adult age, classroom robe, hairstyle and fixed seat, with the specified restrained performance.
<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - sharp original facial identity, clothes, beads and same left-hand open book; only expression and right-hand gestures change.
<Subject 3> (appears in [Shot 1] and any wider class view): fully_preserved - preserve previous student's appearance and own seat while sitting and reacting.
<Subject 4> (appears throughout): fully_preserved - same classroom, six existing female students, seat order, physical light source and wardrobe relationships.
<Picture 4> ([Shot 1] opening composition): partially_preserved - retain framing and people, with the teacher's hand and shoulders adjusted to the planned preceding ending rather than replaying its questioning pose.
<Picture 5> (standing dialogue composition): fully_preserved - preserve girl screen-left, teacher screen-right, both clear faces, realistic adult scale and book support.

detailed_description:
Keep refined photographic realism, equally sharp faces, muted teal-gray walls, dark brown timber and local warm window highlights. Preserve the same physical window on screen-right in this reverse view.

[Shot 1] Use <Picture 4>'s standing composition, but begin with <Subject 2>'s right hand already returned to the lower book edge, matching the planned preceding ending. His left hand supports the open book. <Subject 3> sits onto her original cushion while he turns and takes two steps screen-right to screen-left toward the next desk. Widen slightly and track with him, revealing <Subject 1> seated in her plum-brown robe and long braid down her back. Keep <Subject 4>'s fixed seat order, low desks and central aisle. Do not replay the classroom opening or reset his earlier extended palm.

From 1.20 to 3.60 seconds, <Subject 2> (S1) asks, <d>[Chinese] 早起的鸟儿有什么吃？</d> Fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. From 3.60 to 5.80 seconds, <Subject 1> lays her brush on the desk and rises at her own cushion, keeping the braid behind her shoulders. The fourth student settles her sleeves and watches her; visible classmates exchange small knowing glances, believing the subject concerns ordinary life. They stay seated, without interrupting.

[Shot 2] At 00:06.000, cut to <Picture 5>'s standing pair: fifth woman screen-left, teacher screen-right. Her empty hands rest loosely near her waist, eyelids relaxed, lips neutral; she looks young and composed, not ill, elderly or tearful. From 6.20 to 7.50 seconds, <Subject 1> (S2) answers, <d>[Chinese] 有苦吃！</d> Fixed voice: A young adult female mid-low voice with light breathy resonance and a faint soft grain, relaxed connected syllables, restrained pitch movement and an easy unforced pace. Use a mild natural Northeastern Mandarin cadence, calm and accustomed to the situation. Keep the words clear and the vocal age young; no tired old rasp or comic slurring. Speak plainly as if stating an everyday fact. From 7.50 to 8.30 seconds, she closes her lips and gives a slight accepting nod, leaving a clean answer-hold window.

From 8.30 to 11.20 seconds, the teacher draws one deliberate breath, lifting his shoulders slightly as he prepares to lecture. He opens his mouth, notices her unwavering matter-of-fact expression, and cannot find a reply. The breath slowly escapes and his shoulders drop; he closes his mouth again. His right fingertips remain gently against the lower book edge instead of pointing, recoiling or turning pages. She keeps her eyes on him, calm and fully convinced, with no teasing smile. From 11.20 to 13.00 seconds, keep both faces clear and naturally breathing for the editor's reaction beat. Both remain standing, book open in the same left hand and her hands empty. No extra speech, tea cup, ruler, narration, subtitles or generated freeze frames.

overall_soundscape:
Low continuous classroom ambience, soft footsteps, cushion contact and fabric movement. Only the stated question and answer are intelligible speech; the teacher later draws one controlled breath and lets it out without words. No narrator, laugh track, extra answer or independent music cue.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the sixth young adult woman, with bright large almond eyes, a heart-oval face, a neat narrow chin, a half-up coiled high bun with an ornamental gold pin, long hair half down, and peacock-teal brocade with bronze floral trim. Preserve her exact distinct identity from <Picture 1>, with classroom clothing, hair placement and expression from <Picture 5>.
<Subject 2> is the teacher from <Picture 2>, the original sharp identity reference: same adult face, clear eyes, sparse individually resolved mustache and chin beard, long dark hair, black robe, white collar and wooden beads. Preserve his LEFT-hand support of the same open tan book from <Picture 4> and <Picture 5>.
<Subject 3> is the fifth young adult woman with a soft oval face, a long braid down her back and a plum-brown robe with teal trim from <Picture 4>; she sits silently at her existing desk.
<Subject 4> is the classroom and exactly six female students' seating from <Picture 3>. Opening-camera left to right: ink-blue low bun; pale-sage low braids and ribbon; slate-lavender low chignon; charcoal-olive high bun; plum-brown long braid; peacock-teal half-up long hair. The reverse dialogue camera sees this order reversed. Preserve that geometry, low desks, cushions, timber and physical windows; do not clone or add pupils.
<Picture 4> is an edited opening-composition reference for [Shot 1], not an actual generated video tail frame. Preserve the pair and left-hand open book, adjusting the right-hand opening pose to the preceding planned ending described below.
<Picture 5> anchors the current woman's standing two-person composition in [Shot 2] and any later return to it, after the preceding student sits, the teacher moves and the current student rises.

summary:
[keyframe completion + reference generation] Generate one 15.00-second classroom comedy round with 4 shots. The previous student sits, the teacher moves along the desk front from screen-right to screen-left and asks, then the current woman rises and earnestly gives her answer. Maintain two readable faces in the answer and shared reaction; the student's certainty and teacher's contrasting response carry the comedy. Identity references are not playback frames, and the room reference does not restart the opening.

retention_analysis:
<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4]): fully_preserved - same face, youthful adult age, classroom robe, hairstyle and fixed seat, with the specified restrained performance.
<Subject 2> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4]): fully_preserved - sharp original facial identity, clothes, beads and same left-hand open book; only expression and right-hand gestures change.
<Subject 3> (appears in [Shot 1] and any wider class view): fully_preserved - preserve previous student's appearance and own seat while sitting and reacting.
<Subject 4> (appears throughout): fully_preserved - same classroom, six existing female students, seat order, physical light source and wardrobe relationships.
<Picture 4> ([Shot 1] opening composition): partially_preserved - retain framing and people, with the teacher's hand and shoulders adjusted to the planned preceding ending rather than replaying its questioning pose.
<Picture 5> (standing dialogue composition): fully_preserved - preserve girl screen-left, teacher screen-right, both clear faces, realistic adult scale and book support.

detailed_description:
Preserve cinematic photographic realism, equally sharp adult faces, muted teal-gray walls, brown timber and warm screen-right window light. Keep costume colors stable and avoid beauty-filter smoothing.

[Shot 1] Use <Picture 4>'s standing pair: <Subject 3> screen-left, <Subject 2> screen-right, shoulders relaxed, left hand supporting his open book and right fingertips at its lower edge. She sits onto her own cushion. He composes himself, turns and takes two steps screen-right to screen-left along the desk front toward <Subject 1>. Widen and track left, retaining <Subject 4>'s six seats. No return to the dais, seat swapping or sudden relocation.

From 1.20 to 3.80 seconds, <Subject 2> (S1) asks, <d>[Chinese] 三十六计第一计是什么？</d> Fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. From 3.80 to 5.80 seconds, <Subject 1> puts her brush down and rises in place. Her half-up bun, long loose hair and teal robe remain unchanged. The teacher stops beside her, lifting his right palm near his lower chest in restrained patience. The fifth student watches from her cushion; classmates lean their attention toward the final answer.

[Shot 2] At 00:06.000, cut to <Picture 5>'s standing two-person framing, sixth student screen-left and teacher screen-right. From 6.20 to 8.40 seconds, <Subject 1> (S2) answers, <d>[Chinese] 食品添加剂！</d> Fixed voice: A young adult female upper-middle voice with focused bright resonance and a subtle natural nasal ring, crisp evenly spaced syllables and compact assured pacing. Use a mild natural Northeastern Mandarin cadence, precise and academically certain. Preserve this focused ringing timbre; no breathiness, baby voice, shriek or theatrical elongation. Give all five syllables a complete clear pronunciation; do not swallow the final syllable. She is academically certain, eyes bright, without a knowing joke. From 8.40 to 9.20 seconds, her lips close into a tiny proud expectant smile. The teacher's right hand stalls in midair, eyebrows lift, eyes fix on her and lips part slightly, beginning his speechless reaction.

[Shot 3] At 00:09.200, cut wider from the SAME camera side to the fixed row. The other five women remain seated, smiling and nodding approval. After setting brushes down, they give brief natural approving claps, convinced the answer is correct. Preserve six distinct women and their existing clothes and hairstyles, without additional pupils. Their response is pleased agreement, not mockery or a canned comedy audience.

[Shot 4] At 00:10.800, return to the same pair. The teacher remains caught mid-gesture while she raises her brows a fraction, waiting seriously for praise. Claps fade by 12.00 seconds. From 13.00 to 15.00 seconds, retain this contrasting reaction with subtle breathing, not a literal frozen image or stone transformation. Keep his open book in the left hand and right palm suspended; she stays standing with empty hands. No extra dialogue, narrator, subtitles, lottery epilogue or additional camera move.

overall_soundscape:
Quiet classroom ambience beneath the two specified lines, with soft footsteps and clothing rustles. After the final syllable, five seated classmates give a brief natural burst of approving claps and nonverbal pleased murmurs, then settle; no extra intelligible words or canned cheering. The teacher is speechless, with no narrator, laugh track or magical transformation sound.

non_diegetic_music:
N/A

subject_definitions:
<Subject 1> is the teacher, whose sharp facial identity comes from <Picture 1>: young adult man, long dark hair, sparse moustache and beard, black patterned robe, white collar and wooden prayer beads. Use <Picture 3>, <Picture 4> and <Picture 5> for classroom costume, the SAME open book and staging. He is the only speaking character, T01 (S1).
<Subject 2> is the first young adult woman, with identity from <Picture 2>, ink-blue bronze-floral classroom robe, braided low bun and small flower ornament, as shown in <Picture 4> and <Picture 5>. She is seated initially, rises in Shot 2 and remains silent.
<Subject 3> is the sixth young adult woman in <Picture 3>, wearing peacock-teal brocade with a half-up bun and long loose hair. She stands at the beginning, then sits at her original seat when the teacher leaves; she is silent.
<Subject 4> is the classroom and fixed seated row shown in <Picture 3> and <Picture 4>: dark timber, teal-gray walls, low desks, cushions and the same physical window on screen-right in this reverse camera view. In <Picture 4>, the visible third, second and first women appear left to right in slate-lavender, pale sage and ink-blue robes. The other seats remain offscreen to the left. Exactly six women belong to the class; do not clone or rearrange them.
<Picture 3> is the edited opening composition anchor for Shot 1, reused from the sixth-round storyboard; match the preceding planned stunned pose, not the earlier questioning expression. It is not an actual video-tail frame.
<Picture 4> is the mid-return movement composition anchor for Shot 2, showing the teacher walking toward the first woman along the desk-front lane. Reach it through visible movement after departure, then show her rising.
<Picture 5> is the composition anchor for Shot 3, after arrival and rising: teacher on the left, first woman on the right, clear opposing three-quarter faces and open book in his left hand.

summary:
[keyframe completion + reference generation] Create one 15-second classroom epilogue in three shots. <Subject 1> reprimands the class beside <Subject 3>, pauses silently for two seconds, walks back along <Subject 4>'s fixed row and quietly asks <Subject 2> about the lottery. Use <Picture 3>, <Picture 4> and <Picture 5> in that shot order, preserving the intervening movement and rise. The women have no dialogue.

retention_analysis:
Fully preserve <Subject 1>'s facial identity from <Picture 1> and his costume, book hand and established T01 voice across all three shots; change only expression, gesture and vocal intensity. Preserve <Subject 2>'s identity from <Picture 2>, with clothing and hair fixed by the approved classroom images. Reuse <Subject 3> only for the opening and departure; preserve <Subject 4>'s seat order, lighting and furniture across camera changes. Use <Picture 3> as an edited opening anchor, <Picture 4> as a mid-walk waypoint and <Picture 5> as the final standing two-shot. Do not turn identity references into slideshow shots or morph the sixth woman into the first. No actual video or voice recording is supplied; audible voice continuity remains text-guided.

detailed_description:
Preserve sharp natural faces, stable adult proportions, muted teal-gray walls, dark timber and warm window light from screen-right. Keep the open book supported by the teacher's LEFT hand throughout.

[Shot 1] Start from <Picture 3>'s standing pair, <Subject 3> screen-left and <Subject 1> screen-right. Adapt his expression to the preceding planned ending: lips slightly apart, eyes arrested, right palm suspended. This is a storyboard continuation, not a supplied video-tail frame. He regains speech and looks across the class. From 0.20 to 3.00 seconds, <Subject 1> (S1) says with a slight tremble, <d>[Chinese] 你们满脑子都是什么歪理？</d> Fixed voice: An adult male low-mid baritone, grounded chest resonance with a lightly dry grain, clear consonants, an unhurried conversational pace and a mild natural Northeastern Mandarin cadence. Keep the same vocal age, pitch center, resonance and grain across every round; only emotional intensity changes. No elderly rasp, cartoon growl or announcer delivery. The sixth woman listens seriously, hands empty; she gives no answer. From 3.00 to 5.00 seconds, maintain exactly two seconds without speech or walking. His raised right hand slowly lowers as his eyes turn toward the first student's position; retain breathing and quiet room tone.

[Shot 2] At 00:05.000, cut wider from the same side, showing his turn and first departing step before tracking screen-left to screen-right along the clear lane in front of the desks. <Subject 3> sits onto her own cushion. Preserve <Subject 4>'s seat order and physical window; do not mirror the room or reset him at the dais. Maintain continuous brisk, natural steps, book open in his left hand, right arm relaxed. The camera follows him until <Picture 4>'s mid-return composition is reached: third, second and first women sit in that left-to-right order. The remaining seats continue outside the left edge. This image is a waypoint, not an instant relocation at the cut. Nearby women follow him with their eyes, staying seated. From 7.80 to 9.40 seconds, <Subject 2> puts down anything held and rises at her own cushion; show the rise before changing shots. He arrives beside her by 9.60 seconds and rests his right fingertips against the book edge. Footsteps stop; sleeve rustles subside.

[Shot 3] At 00:09.600, cut to <Picture 5>'s standing pair, teacher screen-left, first woman screen-right, both faces clearly focused. From 9.70 to 10.30 seconds, <Subject 1> (S1) says with restrained embarrassment, <d>[Chinese] 那啥！</d> He inclines slightly toward her, without touching. From 10.50 to 13.30 seconds, <Subject 1> (S1) lowers his volume confidentially and asks, <d>[Chinese] 那……双色球怎么买？</d> Preserve T01's timbre and intelligibility; complete the final syllable before the reaction. Through 15.00 seconds, she silently raises one eyebrow and gives him a small understanding look; he waits expectantly. Keep both standing and naturally breathing. No additional dialogue, narration, subtitles, generated freeze frames, voice replacement or music.

overall_soundscape:
Low continuous classroom room tone, soft cloth and book handling, natural footsteps during the return, then quiet breathing. Only the three specified teacher lines are intelligible speech, all by the same T01 speaker S1. Silence from 3.00 to 5.00 seconds means no speech or music, not a loss of room ambience. Keep the final quiet question audible and complete; do not mute the final two-shot. No female dialogue, scolding improvisation, laughter track or narrator. The lower voice in Shot 3 changes volume and intent, not speaker identity or vocal age.

non_diegetic_music:
N/A