Style Presets

the prompt

【Reference Material Instructions】@Image 3: Scene visual style (street/night/underground space, oppressive composition, environmental layers, film grain) @Image 2: Text packaging style (font texture, graphic design, dynamic graphic typography impact) @Image 1: Character appearance (face, hairstyle, clothing silhouette, proportions, demeanor, overall atmosphere) Only reference the specified dimensions, do not directly copy the reference images, and do not include real brands, logos/titles/recognizable text from the original reference images.
【Core Concept】10 seconds, 16:9 horizontal trap MV. Two fly detective brothers take turns rapping to the camera in multiple close-up underground spaces, following the trap beat rhythm throughout, with hard cuts on the beat, and high-contrast printed poster-style English block text appearing on screen with bass hits. Underground music video + fashion magazine collage + high-fashion texture, a brotherly duo cold performance.
【Visual Process Description】Shot 1 — Extreme close-up on face / Corridor oppression- Scene: Narrow corridor or underground entrance close-up, referencing @Image 3- Shot: Extreme close-up on face (face/eyes/shoulders & neck/clothing collar detail)- Character: Detective A looks directly at the camera and starts rapping, with a calm and sharp gaze- Text: "TWO FLY" huge bold English text presses into the top and bottom of the frame, not blocking the eyes- Rhythm: 808 bass hits, "TWO FLY" instantly compresses vertically and rebounds; hi-hat roll causes fine, rapid vibrations on the letter edges; scanning misalignment on the "fly" downbeat- Hard cut
Shot 2 — Medium close-up half-body / Wall text background- Scene: Another wall/poster wall close-up (clearly different space from Shot 1)- Shot: Clearly pulls out to a medium close-up half-body shot- Character: Detective B raps to the camera, shoulders and head nodding to the hi-hat hits, body slightly leaning forward- Text: "CLUES" huge compressed English text behind the character, naturally occluded by hair/shoulders/clothing silhouette- Rhythm: Text stretches vertically like a poster standing up; with each snare hit, the text suddenly enlarges, shakes, and has scanning misalignment- Hard cut
Shot 3 — Hand close-up / Occlusion switch- Scene: Metal door edge/railing/garage wall/low ceiling detail- Shot: Hand close-up (hand/ring/cuff/clothing material as the main subject)- Character: Detective A's hand gesture pushes towards the camera (handing a clue), face blurred or partially visible in the background- Text: Vertical "BROTHERS" on one side of the foreground, briefly occluded by the hand- Rhythm: When "move as one" is said, the text is pulled into a long vertical strip, hard cut and reassembled with the bass hit- Hard cut
Shot 4 — Shoulder & neck close-up to face close-up / Staircase corner- Scene: Staircase corner/doorway shadow/low corridor- Shot: Shoulder, neck, and collar material close-up → sudden short push-in to face close-up on the snare- Character: Detective B turns head closer to the camera and continues rapping- Text: "AS ONE" bursts out like a magazine headline with the sound, suddenly enlarging on the drum hits- Rhythm: Hard cut repositioning of text during the push-in, never blocking the eyes- Hard cut
Shot 5 — Half-body close-up motion trail / Garage or concrete background- Scene: Garage-like space or concrete background- Shot: Detective A half-body close-up- Character: Main layer with clear lip-sync, secondary layer with dynamic motion trails (photocopier misalignment/frame delay, face not distorted), hand gesture sweeps horizontally across the frame- Text: Vertical "CUT GOLD" appears in the background with jump-cut flashes- Rhythm: "CUT" instantly slices the frame horizontally triggering a flash white hard cut; "GOLD" compresses into a heavy block text then suddenly rebounds- Hard cut
Shot 6 — Dual close-up split screen / Color block division- Scene: Underground entrance/wall/railing/door frame (clearly different space from Shot 5)- Shot: Screen split into two/three zones by color blocks, two detectives' face close-ups and half-body details rapidly hard cut and alternate- Character: Two detectives perform alternately on left and right- Text: "CASE COMPLETE" bold short phrase appears in the center- Rhythm: Hard cut reassembly with each 808 bass hit, letters first flatten then instantly stretch; text fragments jump rapidly during hi-hat roll; finally, scanning misalignment and paper tear-like vibrations appear on the final downbeat- Hard cut
Final — Multi-scene close-up performance montage- Scene: Narrow corridor face close-up, wall medium close-up, hand close-up, staircase corner shoulder & neck close-up, garage half-body detail, doorway shadow dual close-up rapidly alternating- Shot: No full-body shots, no multi-person wide shots, only use close-ups/medium close-ups/face close-ups/hand close-ups/shoulder & neck close-ups/half-body shots- Character: Two detectives take turns rapping to the camera, lip-sync/jaw/breathing/eyebrows & eyes/hand gestures precisely synced to vocal, snare, hi-hat roll, bass hit- Text: Final main text "CASE CLOSED" huge, pressing into the frame- Rhythm: Each bass hit triggers a scene hard cut or shot hard cut; each snare hit triggers text pressing on screen or layer misalignment; on the final beat, all scene fragments freeze synchronously, hard cut to black
【Overall Requirements Supplement】
▍Visual Style (Throughout)- Coarse grain, slight film jitter, photocopy paper grain, halftone dots, print rough edges, scanning misalignment, jump-cut motion trails, high-speed layer misalignment- Editing rhythm extremely fast, only use hard cuts, jump cuts, beat cuts, occlusion cuts, flash white hard cuts, text press screen hard cuts- No fades, no soft transitions
▍Text Packaging (Throughout)- Text is an integral part of the frame, with real layer relationships- Text can appear between foreground/midground/background of the character- Characters can occlude text, text can also press over character details, but must not block eyes and main facial expressions- Text appears/vibrates/stretches/compresses/misaligns/tears/hard cut reassembles following vocal downbeats, snares, hi-hat rolls, 808 bass hits- Style: Heavy compressed sans-serif, huge English block text, vertical stretch, horizontal compression, vertical English arrangement, curved titles, high contrast black/white/red, photocopy paper grain, halftone dots, print damage, scanning misalignment, zine collage feel- Text quantity restrained: Only 1 main text element per shot, at most a small number; no dense small text, no logos
▍Rhythm Rules (Throughout)- hi-hat roll → rapid micro-vibrations, jump-cuts, text fine fragmentation and reassembly- snare → text suddenly enlarges, frame hard cut, character shoulders press down- 808 bass hit → low-frequency screen press, brief frame distortion, text vertical stretch or horizontal compression- vocal keywords → lip-sync, jaw movement, head nodding, hand gestures pushing forward- Character movement, text animation, scene switching, shot switching must all hit the beat together
▍Scene Switching (Throughout)- Multiple close-up scenes rapid hard cuts- Scene atmosphere unified referencing @Image 3, but each shot's space must be clearly different: narrow corridor, wall close-up, doorway shadow, metal/concrete background, light edges, poster wall, staircase corner, low ceiling, garage-like space, underground entrance- No wide establishing shots, no complex large-scale scenes- Each switch triggered by bass hit, snare, vocal downbeat, or text press screen
▍Shot Switching (Throughout)- Changes must be very distinct: extreme face close-up → medium close-up half-body → hand close-up → shoulder & neck close-up → dual close-up split screen → face close-up → half-body detail- Do not use the same shot type consecutively- Each hard cut creates a clear change in camera distance/character orientation/background space/text layering- Camera can have slight handheld feel, sudden push-in, short pan, rapid zoom compression, close-up sway- Do not become soft slow motion
▍Character Rules (Throughout)- Roles: Two fly detectives / brothers- Maintain the face, hairstyle, clothing silhouette, proportions, demeanor, overall atmosphere from @Image 1 character appearance- Realistic matte skin, clean and high-end, natural pores, slight skin texture- No oily AI beauty face- Brotherly duo taking turns rapping, strong performance to the camera
▍Character Performance (Throughout)- Voice cold, aggressive, relaxed but oppressive, rhythm tight, broken, strong beats- Not static posing, it's a performance- Clear lip-sync: lip shape, jaw, expression, breathing follow the vocal- Beat-synced actions: head nodding, shoulder pressing down, hand gestures pushing forward, body leaning forward, turning head, turning towards the camera, finger pointing at the camera
▍Important Restrictions (Throughout)- No full-body shots- No multi-person wide shots- No complex crowd large-scale scenes- Entire video uses only close-ups/medium close-ups/face close-ups/hand close-ups/shoulder & neck close-ups/half-body shots- Only show upper body, face, hand gestures, clothing material, expressions, lip-sync, text packaging- No need to see complete body and feet- No real brands, logos/titles/recognizable original text from the reference images appear