Tags / Machine learning /
Computer vision library
A general library or model for image and video analysis; for one task use object-detection, image-segmentation, ocr, or face-recognition.
Repositories
22
Net stars gained, last 7 days
+1,580
Top-level Tag
Net stars gained each day, last 90 days
Peak days
- 1
Sep 15, 2026 +876 net stars (median day +161)
- Taichu-AI/ZDTaichu5.0-9B+550 (63%)
Created on Sep 3, 2026
- roboflow/supervision+228 (26%)
- google-ai-edge/mediapipe+19 (2%)
- 2
Sep 30, 2026 +564 net stars (median day +161)
- Taichu-AI/ZDTaichu5.0-9B+406 (72%)
- PSRben/VisionHOPE+64 (11%)
Published weights-v1 on Sep 29, 2026 · Created on Sep 27, 2026
- google-ai-edge/mediapipe+14 (2%)
- 3
Oct 1, 2026 +574 net stars (median day +161)
- Taichu-AI/ZDTaichu5.0-9B+404 (70%)
- PSRben/VisionHOPE+66 (11%)
Created on Sep 27, 2026
- Robbyant/lingbot-map+33 (6%)
- 4
Oct 9, 2026 +434 net stars (median day +161)
- Robbyant/lingbot-map+313 (72%)
- opencv/opencv+18 (4%)
- xinntao/Real-ESRGAN+16 (4%)
- 5
Oct 23, 2026 +513 net stars (median day +161)
- PSRben/VisionHOPE+135 (26%)
- kidous2333/SABER+91 (18%)
- Tencent/WeVisOne-VL+85 (17%)
Created on Oct 22, 2026
A peak day has at least 2 times the net stars of the median day of this chart, and at least 100 more. For each peak day, the list shows the repositories that gained the most net stars that day. It also shows their GitHub releases of that day or the day before (UTC), and their creation date when it was 14 days or less before. These events occurred near the same day. They do not show why the stars came.
Largest
Most stars in total now.
description:
Open Source Computer Vision Library
readme:
We are your essential toolkit for computer vision.
readme:
MediaPipe Solutions provides a suite of libraries and tools for you to quickly apply artificial intelligence (AI) and machine learning (ML) techniques in your applications.
description:
Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
description:
The largest collection of PyTorch image encoders / backbones.
- openai/CLIP35K
readme:
CLIP (Contrastive Language-Image Pre-Training) is a neural network trained on a variety of (image, text) pairs.
description:
Real-time multi-person keypoint detection library for body, face, hands, and foot estimation
readme:
**Vision-Language**
description:
Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
readme:
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
Newest Admissions
Entered the index by the rising rule, newest first.
readme:
用于工业图像异常检测与缺陷定位
readme:
ZDTaichu5.0-9B is a multimodal foundation model for general visual understanding, spatial reasoning, agentic tool use, and embodied-AI research.
readme:
a vision-language foundation model that unifies **object-level** perception (object detection, referring localization, counting, pointing, and OCR)
readme:
memfasilitasi perancangan, pengujian, dan penerapan alur kerja *computer vision* tanpa kode secara visual
readme:
can you estimate where each shopper looks in a store using only an ordinary security camera and open source AI models?
readme:
converts a single overhead video stream into identity-resolved behavioral measurements
readme:
100% pure-Rust, zero-FFI computer vision, stereo reconstruction, and mechanistic interpretability engine
readme:
The hierarchical **Tiny / Small / Base** backbones support classification and dense prediction.
How a repository gets this Tag
An agent read the description, the GitHub topics, and the start of the README of each repository, and chose Tags only from the checked-in Tag list. For each Tag it quoted the text, and a code check kept the Tag only when the quote is in that text. The quote under each repository is that evidence. A Tag does not rank repositories and does not say that one can replace another. A wrong Tag? Use "Report a problem with this page" on the page of the repository. Withheld Repositories are left out of all numbers on this page.