The doc issue I have a custom model trained on my own custom AVA format dataset for spatiotemporal activity detection with the config detection/ava/slowfast_kinetics_pretrained_r50_4x16x1_20e_ava_rgb.py. While inferencing, the model is using faster_rcnn_r50_fpn_2x_coco.py to get the person detection...

What is the problem this feature will solve? when i train a BMN, can i do not concat rbg and flow feature? i have got rgb_feat What is the feature? thank you What alternatives have you considered? No response

Branch 0.x branch (0.x version, such as v0.24.1) Prerequisite I have searched Issues and Discussions but cannot get the expected help. I have read the documentation but cannot get the expected help. The bug has not been fixed in the latest version. Environment sys.platform: linux Python: 3.8.10...

The doc issue How can I convert poseC3D's torch to onnx? Every time I use export_onnx_stdet.py, it reports an error " 'Recognizer3D' object has no attribute 'roi_head' ", but recognizer3D indeed does not have roi_head. I tried to export onnx directly in runner.py, but found that the exported onnx is...

Branch 0.x branch (0.x version, such as v0.24.1) Prerequisite I have searched Issues and Discussions but cannot get the expected help. I have read the documentation but cannot get the expected help. The bug has not been fixed in the latest version. Environment pytorch==1.13.1 torchvision==0.14....

题目名称 在 MMAction2 中实现时序动作分割算法 UVAST 题目描述 项目介绍 MMAction2 是一个基于 PyTorch 的开源工具箱，支持众多视频理解模型，包括动作识别、基于骨架的动作识别、时空动作检测和时间动作定位。 此外，它支持广泛使用的学术数据集，并提供许多有用的工具，帮助用户探索模型和数据集，以及实现高质量的算法...

题目名称 实现视频多模态 demo，输入为一段短视频（小于 60s）和用户提问，输出为根据视频内容对用户的回答 题目描述 项目介绍 MMAction2 是一个基于 PyTorch 的开源工具箱，支持众多视频理解模型，包括动作识别、基于骨架的动作识别、时空动作检测和时间动作定位。 此外，它支持广泛使用的学术数据集，并提供许多有用的工...

What is the problem this feature will solve? no What is the feature? no What alternatives have you considered? how to train on uniformer not uniformerv2, there is no train pipeline in configs/recognition/uniformer/uniformer-base_imagenet1k-pre_16x4x1_kinetics400-rgb.py

The doc issue 如果我修改了部分源码，比如loss函数，需要对整个项目再次安装吗 Suggest a potential alternative/fix No response

Branch main branch (1.x version, such as v1.0.0, or dev-1.x branch) Prerequisite I have searched Issues and Discussions but cannot get the expected help. I have read the documentation but cannot get the expected help. The bug has not been fixed in the latest version. Environment ubuntu20.04 cud...