urceolate-genusophioglossum435/awesome-human-activity-recognition
🤖 Explore curated resources for Human Activity Recognition (HAR), including datasets for action recognition, motion capture, and pose estimation.
What's novel
🤖 Explore curated resources for Human Activity Recognition (HAR), including datasets for action recognition, motion capture, and pose estimation.
Code Analysis
13 files read · 3 roundsA curated documentation repository containing ~17 Markdown dataset cards for Human Activity Recognition benchmarks, organized by modality (vision, wearable, skeleton, multimodal, emerging), with minimal Python tooling for catalog generation and ASCII normalization.
Strengths
The dataset cards are well-structured, consistent, and genuinely useful for researchers — each includes modality, scale, license, benchmarks with citations, tooling links, and known challenges. The i18n translations are substantive and the contributing guidelines are thorough.
Weaknesses
The README is misleading — it frames a documentation repo as downloadable 'software' with system requirements and install steps, and claims 'over 40 datasets' when only ~17 exist. There is zero test coverage, no actual ML implementation, and the roadmap promises features (notebooks, reproducibility scripts, web explorer) that don't exist.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
9
Contributors
1
Files
44
Active weeks
3
Repository
Language
Python
Stars
1
Forks
0
License
—