Personal Yoga Pose Video · عوامی Open Skill
Personal Yoga Pose Video
Personal Yoga Pose Video Digen پر ایک عوامی Open Skill ہے۔ کھولیں اور دیکھیں یہ ورک فلو کیا تخلیق کرتا ہے۔
Sports
3 0 1
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4164563/2026-10-11/nvhbsfgd.jpg
# Personal Yoga Pose Video ## When to use - The user uploads a photo of themselves (image 1) and a yoga pose reference image (image 2) and wants a ~30-second video of them doing that pose. - The user wants to redo or tweak the pose, the duration, the music, or the final mixed video. ## Steps 1. First use `write_todos` to list the steps and update the whole list as you progress. 2. Call `recognize_image` on BOTH uploaded images to capture the details: image 1 = the person (face, hair, body type, outfit, setting), image 2 = the yoga pose (exact limb positions, mat, setting, outfit). Keep these descriptions to build the prompt. 3. Generate the yoga video with `generate_video` using `model='ref2v'` (reference-to-video), passing BOTH images as `input_urls` (the person photo + the pose reference) so the person's identity is preserved while they perform the pose. Write the prompt in English describing the person (from image 1) performing the yoga pose (from image 2) with slow, fluid, graceful movement and soft cinematic lighting. Use `orientation='portrait'` (9:16) and `resolution='1080P'` or higher to consume a larger generation budget. Generate 3 clips of ~10 seconds each, each showing the pose from a slightly different angle/flow (e.g. settling into the pose, holding it with a gentle breath, flowing out of it). Name each clip `yoga-clip-1` / `yoga-clip-2` / `yoga-clip-3` with `entity_type='storyboard_videos'`. 4. Generate a calm background music track with `generate_music` (soft ambient / meditative yoga music, instrumental). Name it `yoga-music` with `entity_type='music'`. 5. Mix the 3 clips into a single ~30-second video with `mix_clips` (concatenate the clips in order, add the music track as the audio). Name it `yoga-pose-final` with `entity_type='final_videos'` and `orientation='portrait'`. 6. Reply summarizing the final video in natural language and offer to redo/tweak the pose, duration, or music. ## Notes - Video generation does not support NSFW/adult content on any channel — politely decline such requests. - Use high resolution and multiple clips to consume a larger generation budget (the user asked for ~150 credits or more). - Intermediate clips will be visible on the canvas; that is expected and not an error. - Report failures honestly and never fabricate URLs. - The skill has a preset cover image bound as `cover_image` (a yoga-themed image); refer to it as the skill's cover/thumbnail when needed.