オープンソースの素晴らしい動画プロンプト集。 · 3672 個の素晴らしいプロンプトを収集しました
I was reading about density mapping tonight and coincidentally saw a post talking about how important it is for models to be grounded in real world physics. Now, while that is true, the images can still come out messy if the prompt doesn't control where the detail should live. So, instead of only prompting: "highly detailed environment" I'm trying to structure the scene by detail zones as to not put too much visual information everywhere. As usual (lately), this is 100% overkill and not needed for casual generations Basic structure: High density → where the viewer should look first Medium-high density → where the scene needs grounding or interaction Medium density → supporting environment / framing elements Low density → areas that need readability or breathing room Atmospheric density → far background, haze, sky, distance, soft storytelling Silhouette zone → clean space behind characters or important objects Simple formula: Highest detail concentrated around [main focal/action area]. Medium-high detail placed on [interaction zone / floor / props]. Medium detail used for [side structures / framing / worldbuilding]. Low detail kept behind [characters / subject / text / important shapes]. Far background simplified with [haze / gradients / large readable forms]. Avoid equal detail everywhere so the image has clear visual hierarchy. For this test, I used it to create a futuristic fighting environment for two sci-fi characters and since this was meant for a fight scene, the environment needed a readable combat lane, clear ground contact room for movement, open space behind the fighters... so the prompt focused details around the arena platform, floor seams, rails, and side architecture, while keeping the center space and character silhouette areas cleaner. All prompts below!
I was reading about density mapping tonight and coincidentally saw a post talking about how important it is for models to be grounded in real world physics. Now, while that is true, the images can still come out messy if the prompt doesn’t control where the detail should live.
So, instead of only prompting: “highly detailed environment” I’m trying to structure the scene by detail zones as to not put too much visual information everywhere.
As usual (lately), this is 100% overkill and not needed for casual generations
Basic structure: High density → where the viewer should look first Medium-high density → where the scene needs grounding or interaction Medium density → supporting environment / framing elements Low density → areas that need readability or breathing room Atmospheric density → far background, haze, sky, distance, soft storytelling Silhouette zone → clean space behind characters or important objects
Simple formula: Highest detail concentrated around [main focal/action area]. Medium-high detail placed on [interaction zone / floor / props]. Medium detail used for [side structures / framing / worldbuilding]. Low detail kept behind [characters / subject / text / important shapes]. Far background simplified with [haze / gradients / large readable forms]. Avoid equal detail everywhere so the image has clear visual hierarchy.
For this test, I used it to create a futuristic fighting environment for two sci-fi characters and since this was meant for a fight scene, the environment needed a readable combat lane, clear ground contact room for movement, open space behind the fighters… so the prompt focused details around the arena platform, floor seams, rails, and side architecture, while keeping the center space and character silhouette areas cleaner.
All prompts below!
Kling 3.0Extreme close-up on [SUBJECT]'s eye, [EYE DETAIL]. The camera holds for [DURATION]. Then a slow, [PACE] pull back begins. The face emerges — [FACIAL EXPRESSION]. The pull back continues [MOVEMENT DETAIL], revealing [BODY/CONTEXT]. The camera keeps pulling back to finally reveal [THE TWIST/REVEAL]. [TECHNICAL SPECS].
Seedance 2.0A shaky, panicked handheld recording from the balcony of a sea-facing hotel during the day. The person filming suddenly spots something in the distance, does a jerky zoom-in, then quickly refocuses as a massive giant meteor streaks across the sky and crashes into the ocean with enormous force. The impact creates a colossal tsunami that rushes toward the shore. The wave violently annihilates the beach, shacks, and entire coastline in a brutal display of nature's power. The person filming is in complete panic — camera extremely shaky, breathing heavy. Water surges dramatically, rising rapidly until it reaches just a few floors below the balcony. A water droplet splashes onto the camera lens. The building rocks and shakes violently from the shockwave. In the background, people are screaming in terror while emergency sirens blare with back-and-forth intensity. Cinematic, hyper-realistic, intense documentary-style footage, natural daylight, high tension, raw and chaotic energy.
higgsfieldThe Higgsfield supercomputer running on Gemini is actually wild. The text generation feels way more detailed and coherent, the cinematic motion quality looks insane, and the frame-by-frame control gives creators crazy precision. Even the search feels smarter with deep global knowledge behind it. AI video creation is evolving fast.
Seedance 2.0Ultra-cinematic cyber-occult sequence inside a hidden Tokyo shrine buried beneath massive megacity highways and industrial infrastructure at night. A vast underground Shinto sanctuary exists beneath layers of elevated expressways, train lines, pipes, cables, and neon-lit maintenance corridors. Ancient torii gates stand surrounded by concrete pillars covered in condensation, holographic signage flickers beside weathered prayer tablets, and sacred lanterns glow softly beneath brutalist architecture. The camera slowly descends from rain-soaked elevated highways into hidden underground access tunnels illuminated by flickering red emergency lights. Steam rises from vents while distant trains rumble overhead, causing dust and hanging prayer charms to vibrate subtly. Massive highway support columns tower above the shrine complex like cathedral pillars. Neon reflections ripple across flooded stone pathways. Sacred ropes, paper talismans, and old wooden prayer plaques sway gently in warm underground wind currents. Monks in dark ceremonial robes move silently through the shrine carrying glowing lanterns while holographic spirits flicker briefly in peripheral darkness. Ancient statues emerge from shadow beside exposed wiring and industrial pipes leaking steam. The camera glides continuously through narrow corridors, hidden staircases, flooded maintenance tunnels, and finally into the main shrine chamber where an enormous sacred tree grows through cracked concrete and steel foundations. Lightning flashes from openings far above the megacity ceiling, briefly illuminating drifting incense smoke and thousands of hanging prayer seals. Soft bioluminescent spirits move through the air like glowing embers. Final cinematic reveal: the hidden shrine rests directly beneath roaring multilayer expressways while rivers of traffic streak overhead like flowing stars, unaware that an ancient spiritual sanctuary survives below the city. Style: ultra cinematic cyberpunk spiritualism, hidden urban shrine aesthetic, dark futuristic Tokyo atmosphere, realistic wet surfaces, volumetric fog, neon reflections, industrial decay mixed with sacred architecture, smooth floating cinematography, grounded environmental realism, atmospheric anime-film quality, no text, no subtitles, no overlays.
Seedance 2.0Create a ultra cinematic realistic WWII war drama video that exactly follows the 8-panel storyboard in the reference image, Hollywood movie quality like Dunkirk + Saving Private Ryan, photorealistic details, emotional storytelling, powerful directorial shots, smooth cinematic camera movements with minimal hard cuts, 1080p. [0-3s] Panel 1: Majestic wide aerial shot slowly pushing in over a massive intense dogfight above a burning city at dusk, dozens of fighter planes, explosions, smoke and fire everywhere, epic establishing shot with dramatic sky. [3-5.5s] Panel 2: Intense close-up inside the cockpit - Young exhausted pilot wearing goggles and oxygen mask, focused eyes, flying through chaos, camera subtly shakes with the plane, dramatic side lighting. [5.5-8s] Panel 3: Dynamic tracking shot following the pilot's plane in a fierce dogfight, enemy planes shooting, tracers and explosions around, fast but smooth aerial camera movement circling the action. [8-10.5s] Panel 4: Dramatic crash sequence - The damaged plane smoking and spiraling down, hard landing in a rural field, pilot climbing out of the wreckage, camera follows the impact and movement realistically. [10.5-13s] Panel 5: Emotional intimate close-up - The injured pilot sitting inside the wrecked cockpit, slowly reading an old handwritten letter, soft golden light on his tired dirty face, very subtle camera push-in, deep emotion. [13-16s] Panel 6: Tense confrontation shot - The pilot surrounded by German soldiers aiming rifles in a war-torn village, dramatic standoff, slow circling camera building tension, realistic gritty details. [16-19s] Panel 7: Emotional rescue moment - The wounded pilot being helped and carried by resistance fighters (man with beret and woman) in the village, human connection, warm cinematic lighting, smooth tracking movement. [19-22s] Panel 8: Epic final wide shot - The pilot standing with his back to camera on a hill at sunset, watching a formation of allied planes flying across the dramatic orange sky, camera slowly rises up and pulls back, powerful and hopeful ending. #Seedance2 on @yapper_so
Seedance 2.0Dense futuristic megacity at night covered in fog. A woman moves through crowded elevated walkways while emergency broadcasts interrupt giant holographic screens overhead. Suddenly, massive shadows begin moving through the fog between skyscrapers. People stop walking. Then a gigantic creature silhouette emerges partially through the mist: enormous biomechanical legs stepping slowly between buildings. Every footstep shakes the city. Cars alarms activate automatically. Glass trembles. Streetlights flicker. The creature is too massive to fully see at once. Only fragments visible through fog: glowing eyes, giant armored limbs, mechanical breathing vents, cables hanging beneath its body. The woman begins running through packed pedestrian corridors. Behind her, the colossal entity slowly advances through the skyline. Skyscraper lights reflect across its metallic surface. As tension escalates: drones fall from the sky, bridge structures vibrate violently, giant holograms distort, crowds panic and stampede. The creature lowers its massive head between towers searching the streets below. Its glowing eye scans directly toward her position. At climax, the enormous creature steps over the avenue directly behind her. The frame fills with gigantic mechanical limbs and steam. Then suddenly, the creature stops moving. Silence. Fog slowly hides it again. Ultra cinematic sci-fi tension, gigantic scale, grounded realism, atmospheric fog, cyberpunk city visuals, environmental panic, cinematic handheld camera, colossal creature design.