ShiroNekoe/video2robot

🤖 Generate robot motions from video or prompts using an efficient end-to-end pipeline for human pose extraction.

What's novel

🤖 Generate robot motions from video or prompts using an efficient end-to-end pipeline for human pose extraction.

Code Analysis

13 files read · 4 rounds

An orchestration pipeline that chains AI video generation (Veo/Sora), 3D human pose extraction (PromptHMR), and robot motion retargeting (GMR) to produce humanoid robot joint trajectories from text prompts, with a web UI and 3D visualization.

Strengths

Well-structured integration of three complex ML/robotics components with thoughtful prompt engineering for pose-friendly video generation, correct coordinate transforms, multi-track support, and a substantial custom 3D visualization that renders robot meshes alongside source video.

Weaknesses

No tests whatsoever, the README is completely misleading (describes a consumer desktop app when this is a research tool requiring conda envs and API keys), and the core ML logic is entirely delegated to third-party libraries making this primarily a glue/orchestration layer.

Score Breakdown

Innovation

5 (25%)

Craft

47 (35%)

Traction

11 (15%)

Scope

61 (25%)

Signal breakdown

Innovation

Not Fork+1

Code Novelty+1

Concept Novelty+1

Craft

Ci-3

Tests-5

Polish+0

Releases-2

Has License+5

Code Quality+18

Readme Quality+15

Recent Activity+7

Structure Quality+5

Commit Consistency+2

Has Dependency Mgmt+5

Traction

Forks+0

Stars+6

Hn Points+0

Watchers+3

Early Traction+0

Devto Reactions+0

Community Contribs+2

Scope

Commits+3

Languages+8

Subsystems+13

Bloat Penalty+0

Completeness+7

Contributors+6

Authored Files+12

Readme Code Match+3

Architecture Depth+7

Implementation Depth+8

Evidence

Commits

Contributors

Files

Active weeks

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

Forks

License

NOASSERTION