2000 character limit reached
A Vanilla Multi-Task Framework for Dense Visual Prediction Solution to 1st VCL Challenge -- Multi-Task Robustness Track (2402.17319v1)
Published 27 Feb 2024 in cs.CV
Abstract: In this report, we present our solution to the multi-task robustness track of the 1st Visual Continual Learning (VCL) Challenge at ICCV 2023 Workshop. We propose a vanilla framework named UniNet that seamlessly combines various visual perception algorithms into a multi-task model. Specifically, we choose DETR3D, Mask2Former, and BinsFormer for 3D object detection, instance segmentation, and depth estimation tasks, respectively. The final submission is a single model with InternImage-L backbone, and achieves a 49.6 overall score (29.5 Det mAP, 80.3 mTPS, 46.4 Seg mAP, and 7.93 silog) on SHIFT validation set. Besides, we provide some interesting observations in our experiments which may facilitate the development of multi-task learning in dense visual prediction.
- Adabins: Depth estimation using adaptive bins. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4009–4018, 2021.
- nuscenes: A multimodal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020.
- Detrdistill: A universal knowledge distillation framework for detr-families. arXiv preprint arXiv:2211.10156, 2022.
- Deliberated domain bridging for domain adaptive semantic segmentation. Advances in Neural Information Processing Systems, 35:15105–15118, 2022.
- Bevdistill: Cross-modal bev distillation for multi-view 3D object detection. ICLR, pages 1–17, 2022.
- Graph-DETR3D: rethinking overlapping regions for multi-view 3D object detection. In ACM MM, pages 5999–6008, 2022.
- Ddod: Dive deeper into the disentanglement of object detector. IEEE Transactions on Multimedia, 2023.
- Masked-attention mask transformer for universal image segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1290–1299, 2022.
- Exploring recurrent long-term temporal fusion for multi-view 3d perception. arXiv preprint arXiv:2303.05970, 2023.
- Mask r-cnn. In Proceedings of the IEEE international conference on computer vision, pages 2961–2969, 2017.
- Bevdet4d: Exploit temporal cues in multi-camera 3d object detection. arXiv preprint arXiv:2203.17054, 2022.
- 1st place solutions of waymo open dataset challenge 2020–2d object detection track. arXiv preprint arXiv:2008.01365, 2020.
- Pointrend: Image segmentation as rendering. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9799–9808, 2020.
- Depthformer: Exploiting long-range correlation and local information for accurate monocular depth estimation. arXiv preprint arXiv:2203.14211, 2022.
- Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers. In European conference on computer vision, pages 1–18. Springer, 2022.
- Binsformer: Revisiting adaptive bins for monocular depth estimation. arXiv preprint arXiv:2204.00987, 2022.
- Sparse4d: Multi-view 3d object detection with sparse spatial-temporal fusion. arXiv preprint arXiv:2211.10581, 2022.
- Time will tell: New outlooks and a baseline for temporal multi-view 3d object detection. arXiv preprint arXiv:2210.02443, 2022.
- Shift: a synthetic driving dataset for continuous multi-task domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21371–21382, 2022.
- Fcos3d: Fully convolutional one-stage monocular 3d object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 913–922, 2021.
- Detr3d: 3d object detection from multi-view images via 3d-to-2d queries. In Conference on Robot Learning, pages 180–191. PMLR, 2022.
- Stronger, fewer, & superior: Harnessing vision foundation models for domain generalized semantic segmentation. arXiv preprint arXiv:2312.04265, 2023.
- Zehui Chen (41 papers)
- Qiuchen Wang (5 papers)
- Zhenyu Li (120 papers)
- Jiaming Liu (156 papers)
- Shanghang Zhang (173 papers)
- Feng Zhao (110 papers)