This is the official PyTorch implementation of our paper:
Treating Motion as Option to Reduce Motion Dependency in Unsupervised Video Object Segmentation, WACV 2023
Suhwan Cho, Minhyeok Lee, Seunghoon Lee, Chaewon Park, Donghyeong Kim, Sangyoun Lee
Link: [WACV] [arXiv]
Treating Motion as Option with Output Selection for Unsupervised Video Object Segmentation, arXiv 2023
Suhwan Cho, Minhyeok Lee, Jungho Lee, MyeongAh Cho, Sangyoun Lee
Link: [arXiv]
You can also find other related papers at awesome-video-object-segmentation.
In unsupervised VOS, most state-of-the-art methods leverage motion cues obtained from optical flow maps in addition to appearance cues. However, as they are overly dependent on motion cues, which may be unreliable in some cases, they cannot achieve stable prediction. To overcome this limitation, we design a novel motion-as-option network that is not much dependent on motion cues and a collaborative network learning strategy to fully leverage its unique property. Additionally, an adaptive output selection algorithm is proposed to maximize the efficacy of the motion-as-option network at test time.
1. Download DUTS, DAVIS, FBMS, YouTube-Objects, and Long-Videos from the official websites.
2. Estimate and save optical flow maps from the videos using RAFT.
3. For convenience, I also provide the pre-processed DUTS, DAVIS, FBMS, YouTube-Objects, and Long-Videos.
4. Replace dataset paths in "run.py" file with your dataset paths.
1. Open the "run.py" file.
2. Specify the model version.
3. Verify the training settings.
4. Start TMO training!
python run.py --train
1. Open the "run.py" file.
2. Specify the model version and "aos" option.
3. Choose a pre-trained model.
4. Start TMO testing!
python run.py --test
pre-trained model (rn101)
pre-trained model (mitb1)
pre-computed results
Code and models are only available for non-commercial research purposes.
If you have any questions, please feel free to contact me :)
E-mail: suhwanx@gmail.com