-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
demo演示的时候多人不能全部检测出人体姿态 #700
Comments
请更新到最新版的mmpose. |
看起来和这个PR 有关 #671 |
您好,这么多人姿态估计模型整体的性能怎么样啊 |
可以看看 model zoo,里面汇总了所有模型的指标。 |
Also, here is comparison on the same dataset, e.g. coco https://mmpose.readthedocs.io/en/latest/papers/datasets.html#div-align-center-coco-eccv-2014-div |
@innerlee 模型的FPS性能怎么样啊?能达到实时性吗? |
FPS取决于具体的模型和硬件环境. 我们计划添加推理速度的测试工具和benchmark, 可以关注一下近期的更新 |
可以先参考这个来自社区的topdown模型推理速度的demo: #560 (comment) |
@ly015 您好,目前来说这个仓库里面的模型有部署在Iphone上的推理速度达到150FPS的吗? |
目前没有。现在mmpose里面的提供的模型,主要是学术界常用的模型,暂时没考虑直接部署。 |
@jin-s13 您好,这个仓库有考虑做一些模型量化和模型压缩方面的工作吗? |
OpenMMLab有这方面的整体计划, 但短期内mmpose不会单独做这方面的支持 |
Please open separate issues for new topics :) |
…-mmlab#700) * Update distributed_training.py Better example for DDP training * Update distributed_training.py * Update distributed_training.py update according to reviwer's suggesstions. * Update distributed_training.py * Update distributed_training.py The previous update copy data from main branch, its a mistake. This update fix this mistake and the code is tested.
为什么只能检测出来一个人的姿态?是代码问题还是模型问题啊
The text was updated successfully, but these errors were encountered: