sprited/dancing-stick-figures
Dancing Stick Figures — v0.2 A small, fully-labelled synthetic video dataset for learning (and teaching) video diffusion on one consumer GPU. 1,340 clips · 6 s @ 20 fps · 128×128 RGBA · 482,400 frames · 134 text prompts × 10 seeds × 3 cameras · every frame carries the 3D skeleton, camera and G-buffer (depth, normals, part segmentation) that produced it. Think of it as an MNIST for video generation: small enough that a 64² video diffusion model trains from scratch in a few… See the full description on the dataset page: https://huggingface.co/datasets/sprited/dancing-stick-figures.
v0.2 card: curated counts (134 prompts / 1,340 motions / 482,400 frames), curation note
v0.2 curation: remove superseded shards left by the previous partitioning
v0.2 frames: visual-QA prompt curation (143 -> 134 prompts; see prompts/v02_excluded.txt in the code repo)
v0.2 mini: visual-QA prompt curation (143 -> 134 prompts; see prompts/v02_excluded.txt in the code repo)
v0.2 motion: visual-QA prompt curation (143 -> 134 prompts; see prompts/v02_excluded.txt in the code repo)
v0.2 card: seed-disjoint in-domain splits (seeds 0-7/8/9), 120f contract, historical v0.1 baselines marked
Release v0.2 seed-disjoint in-domain splits (part 3)
Release v0.2 seed-disjoint in-domain splits (part 2)
Release v0.2 seed-disjoint in-domain splits
card: image columns first in dataset_info (viewer shows color+text first)
Upload README.md with huggingface_hub
card: video baselines + AR rollout
card: start-here route + Colab
card: mini config
mini config (64px colour+seg+labels) (part 2)
mini config (64px colour+seg+labels)
card: code repo link
card: dataset_info features (image columns)
v0.1 initial upload (part 2)
v0.1 initial upload
initial commit