Jas

icon indicating an email [email protected]

icon company University of Hong Kong icon location Hong Kong

Results 166 comments of Jas

Hrnet augumentation detailed！

1.可以具体看config文件 2.训练的时候，用gt框，不用检测框

Hrnet augumentation detailed！

https://github.com/open-mmlab/mmpose/blob/master/configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/hrnet_w32_coco_256x192.py#L95 这里可以看到，我们训练中用到的所有aug

Hrnet augumentation detailed！

mmpose 没有支持 yolov5，可以去 [mmyolo](https://github.com/open-mmlab/mmyolo) 看一下

Hrnet augumentation detailed！

https://github.com/open-mmlab/mmpose/blob/master/demo/docs/2d_human_pose_demo.md#using-mmdet-for-human-bounding-box-detection mmdet检测+mmpose关键点

[Feature] Support ViTPose

Thanks. Both OneDrive and Google Drive are welcome. BTW, would you mind adding some unit-tests? An example could be found https://github.com/open-mmlab/mmpose/pull/1907/files#diff-dadc2075341a40335f28131ceaf3d0d415e5c316c54bb6b0a0741aeb002db24e

Mismatch between weights and config file.

TCFormer wholebody models are supported in mmpose v0.28.0.

训练速度较慢

@chengmengli06 你可以在下面的链接里找到 ‘Supplemental’ 下载地址 https://papers.nips.cc/paper/7363-connectionist-temporal-classification-with-maximum-entropy-regularization

Error when training

@xwjabc I met possibly the same DAverageMeter error (AccuracyNovel is missing). Could you please tell me how to fix it?

confusing in train processing

Thank you for reporting this issue. I appreciate you taking the time to bring this to our attention. Unfortunately, the original author of this repository has since graduated and taken...

The architecture in the implementation is not consistent with what was described in the paper

Thank you for reporting this issue. I appreciate you taking the time to bring this to our attention. Unfortunately, the original author of this repository has since graduated and taken...