本文共 1173 字,大约阅读时间需要 3 分钟。
视频效果:
研究资料:
Pose estimation
COD20K Dataset contain training and testing data for detection and pose estimation of cars on highways. There is approximately 10000 training images with almost 20000 annotated training car instances. The testing set contains approximately 1100 images with over 3000 instances. Each instance annotation contain its bounding box and 3D viewpoint vector (see for details).
Filename | Description | Size |
---|---|---|
complete pose estimation system | 1259 kB | |
description of contents | 39 kB | |
license | 1 kB |
Filename | Description | Size |
---|---|---|
Description of contents. | 2KB | |
Basic code for detection and pose estimation with pre-trained full-body and upper-body models. | 1MB | |
Full code for training and testing, including BUFFY, PARSE, and INRIA image benchmarks. | 89MB | |
Full code for training and testing, including image benchmarks and new evaluation routines. | 89MB |
本文转自 stock0991 51CTO博客,原文链接:http://blog.51cto.com/qing0991/1888363