zzh

TensorRT C++ inference results do not have bbox with custom dataset

2

I can use author's 'model_trt.engine' deploy 'palace.mp4' normally.  However, I generate 'bytetrack_s.onnx' , with my dataset,and make 'model_trt.engine', there have no bbox when inference. Who can tell me why?...

跟踪的框偏大

10

 yolox-s检测后的框大小正常，bytetrack跟踪后的框就变得很大，这是什么情况呢？

卡尔曼滤波更新边框的tlwh

1

在使用每个 detections[idet] 更新每条track时 ，在def update(self, new_track, frame_id):函数里面，我只看到了卡尔曼滤波更新了self.mean, self.covariance等值，没看到在哪里更新了self.tlwh ？ 为什么在跳出update这个函数后，self.tlwh得到了更新？  

RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR

1

 I meet this problem when running train.py with 'python3 tools/train.py -f exps/example/mot/yolox_x_ablation.py -d 8 -b 48 --fp16 -o -c pretrained/yolox_x.pth' Could you tell me how to solve the problem?

Train loss not decrease

2

I used this code to train with VOC2007 dataset. The train loss decrease from 50 to 9 at 1 epoch. However, train loss maintain 9 at other epoches,and I used...

### 问题确认 Search before asking - [X] 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question I use QualityInspector STFPM with MvTec AD...

我按照‘’PaddleVideo快速开始‘’的步骤，装上paddlepaddle-gpu、ppvideo==2.3.0之后。运行ppvideo --model_name='ppTSM_v2' --use_gpu=False --video_file='data/example.avi'时报错‘’ppvideo: command not found‘’.这怎么解决呢？