msracver / Flow-Guided-Feature-Aggregation

Flow-Guided Feature Aggregation for Video Object Detection
MIT License
723 stars 190 forks source link

the outputs of runing demo.py. #64

Open qiqiang01 opened 4 years ago

qiqiang01 commented 4 years ago

I run the demo.py, freedom from error. The ideal output are the detction images with bbox, category and accuracy. However, the output are the original images without bbox. How can I do? Please, thanks.

H11zang commented 4 years ago

I meet this problem,too. I wanted to detect person ,but the output images without bbox.And I find that only 30 categories without person(you can find the categories in demo.py).You maybe do the same.