Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Official code for MAMMA: Markerless Accurate Multi-person Motion Acquisition.
| Date | Stars |
|---|---|
| 2026-07-24 | 737 |
| 2026-07-25 | 740 |
| 2026-07-28 | 743 |
| 2026-07-30 | 743 |
| 2026-08-06 | 743 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center"> # MAMMA: Markerless Accurate Multi-person Motion Acquisition Hanz Cuevas Velasquez<sup>1\*</sup>, Anastasios Yiannakidis<sup>1\*</sup>, Soyong Shin<sup>2</sup>, Giorgio Becherini<sup>1</sup>, Markus Höschle<sup>1</sup>, Joachim Tesch<sup>1</sup>, Taylor Obersat<sup>1</sup>, Tsvetelina Alexiadis<sup>1</sup>, Eni Halilaj<sup>2</sup>, Michael J. Black<sup>1</sup> <sup>1</sup>Max Planck Institute for Intelligent Systems, Tübingen <sup>2</sup>Carnegie Mellon University <sup>\*</sup>Equal contribution  **[CVPR 2026 Oral]** | [Paper](https://openaccess.thecvf.com/content/CVPR2026/html/Velasquez_MAMMA_Markerless_Accurate_Multi-person_Motion_Acquisition_CVPR_2026_paper.html) | [arXiv](https://arxiv.org/abs/2506.13040) | [Project Page](https://mamma.is.tue.mpg.de/) | [Datasets](https://mamma.is.tue.mpg.de/download.php) </div> ## News - [2026-06] 🎉 MAMMA being presented at CVPR 2026 - [2026-06] Code released (inference + training) --- ## Install ```bash git clone https://github.com/cuevhv/mamma.git cd mamma ``` Full env + CUDA + weights setup: **[docs/INSTALL.md](docs/INSTALL.md)**. ```bash micromamba activate mamma # or: conda activate mamma python -m inference doctor # verify env vars + weight paths ``` The pipeline is zero-config when weights live under `data/`. --- ## Quick demo Bundled 4-cam example, ~56 MB: ```bash bash data/download_example.sh # fetches videos to data/mamma_example/ ``` ```bash python -m inference run \ --cfg configs/examples/presets/quick.yaml \ --footage data/mamma_example \ --seq_name pushing_and_lifting_from_ground \ --calib configs/examples/calib/iphones_outdoors.yaml \ --out-tag demo -v ``` Outputs land under `output/ma_*/demo/mamma_example/…`. **Prefer a browser UI?** Run `bash gui/scripts/dev.sh`, open <http://localhost:3000>, and click **Run demo**. It's the same pipeline but friendlier UX! --- ## Pipeline `ma_cap → ma_masks → ma_2d → ma_3d → ma_vis` | Step | What it does | |------------|----------------------------------------------| | `ma_cap` | Loads multi-view capture | | `ma_masks` | Per-person segmentation (SAM + YOLO) | | `ma_2d` | 2D landmark detection (MammaNet) | | `ma_3d` | Multi-view SMPL-X optimization | | `ma_vis` | Per-camera overlays + interactive scene | Entry point: `python -m inference run` (source: [`inference/cli/run.py`](inference/cli/run.py)). | Argument | What it is | |------------------------|---------------------------------------------------------------------------------------------| | `--cfg` / `--preset` | Pipeline-configuration YAML — declares which steps run and their hyperparameters. Capture-independent. ([what a preset is + how to modify one](docs/CONFIGS.md)) | | `--footage` | Dataset root containing sequence subdirs (use with `--seq_name` + `--calib`). ([layout reference](docs/YOUR-DATA.md#1-lay-out-your-footage)) | | `--seq_name` | One sequence subdirectory name under `--footage` to process (one run = one sequence). | | `--calib` | Calibration file (`.yaml` / `.xcp` / OpenCV `.json`); applies to every sequence under `--footage`. ([format reference](docs/YOUR-DATA.md#2-author-the-calibration-file)) | | `--capture` | Advanced: capture JSON pointing at footage, calibration, sequences, and camera names — used to iterate over many sequences in one invocation. ([schema reference](docs/YOUR-DATA.md#3-mint-a-capture-descriptor)) | | `--out-tag` | Output sub-directory tag under `output/ma_*/<tag>/` (default: `local`). | | `-v` | Verbose runner logs. | ### Run the pipe
Excerpt of 8,723 characters
Read on GitHub
19
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:7b3c79ee10791153, topic:deep-learning, topic:pytorch
matched fp:7b3c79ee10791153, topic:synthetic-data
matched fp:7b3c79ee10791153, topic:computer-vision, topic:pose-estimation