search

Zhongdao / UniTrack

[NeurIPS'21] Unified tracking framework with a single appearance model. It supports Single Object Tracking (SOT), Video Object Segmentation (VOS), Multi-Object Tracking (MOT), Multi-Object Tracking and Segmentation (MOTS), Pose Tracking, Video Instance Segmentation (VIS), and class-agnostic MOT (e.g. TAO dataset).
MIT License
338 stars 34 forks source link

Old YOLO detector Models #21

Open sharoseali opened 2 years ago

sharoseali commented 2 years ago

Can I use previous yolo detector models for MOT (object tracking) like yolov3 /yolov4 networks trained on our custom dataset. What possible changes we have to do if I can modify the existing code for yolov3. THanks

Zhongdao commented 2 years ago

Hi @sharoseali , Yes of course you can replace the YOLOX detector with other detectors. I think you only need to modify demo/demo_mot.py. Please have a look at the code then you will understand :)

sharoseali commented 2 years ago

Hi @sharoseali , Yes of course you can replace the YOLOX detector with other detectors. I think you only need to modify demo/demo_mot.py. Please have a look at the code then you will understand :)

Hi @Zhongdao i have my YOLOv3 custom detection models in .weights file. I want to use them in Multi class Multi tacker. I go through the code but didn't find any code parsing cfg and .weights files used in official dark-net repo. is there any way that i can use them.

Zhongdao commented 2 years ago

Hi @sharoseali, I am not sure I understood your question, maybe you could use code in this repo for parsing cfg and loading model weights?

liuyang77712 commented 2 years ago

Hi @sharoseali , Yes of course you can replace the YOLOX detector with other detectors. I think you only need to modify demo/demo_mot.py. Please have a look at the code then you will understand :)

Hi @Zhongdao i have my YOLOv3 custom detection models in .weights file. I want to use them in Multi class Multi tacker. I go through the code but didn't find any code parsing cfg and .weights files used in official dark-net repo. is there any way that i can use them.

do you solve this problem?

sharoseali commented 2 years ago

Hi @sharoseali , Yes of course you can replace the YOLOX detector with other detectors. I think you only need to modify demo/demo_mot.py. Please have a look at the code then you will understand :)

Hi @Zhongdao i have my YOLOv3 custom detection models in .weights file. I want to use them in Multi class Multi tacker. I go through the code but didn't find any code parsing cfg and .weights files used in official dark-net repo. is there any way that i can use them.

do you solve this problem?

Not yet.... I choose to work on MCMOT YOLOV4. It contains YOLOv3/v4 parsing code but also many problems exist there and needed to be fixed.😐