search

open-mmlab / mmtracking

OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
https://mmtracking.readthedocs.io/en/latest/
Apache License 2.0
3.54k stars 593 forks source link

fine-tune siameserpn++ on my own dataset #617

Open deepqd98 opened 2 years ago

deepqd98 commented 2 years ago

I want to fine-tune siameserpn++ on my own dataset, but my dataset format modified and annotated according to CocoVID's format cannot be used directly. For example, an error is reported there for pairsampling. Do I need to annotate it according to the SOT‘s’ format, and can you give me an example of the sot dataset format and annotation format?

JingweiZhang12 commented 2 years ago

Can you post your error here? About training the mode using your own datasets, you can refer to https://github.com/open-mmlab/mmtracking/blob/master/docs/en/dataset.md