面向第一届VCL挑战赛多任务鲁棒性赛道的稠密视觉预测普通多任务框架解决方案
A Vanilla Multi-Task Framework for Dense Visual Prediction Solution to 1st VCL Challenge -- Multi-Task Robustness Track
AI总结:
本文提出普通多任务框架UniNet,融合DETR3D、Mask2Former和BinsFormer,以InternImage-L为骨干,在SHIFT验证集上取得49.6总分,为稠密视觉预测多任务学习提供参考。
摘要:
本报告介绍了我们在ICCV 2023研讨会第一届视觉持续学习(VCL)挑战赛多任务鲁棒性赛道上的解决方案。我们提出了一个名为UniNet的普通框架,将多种视觉感知算法无缝结合到一个多任务模型中。具体而言,我们分别选择DETR3D、Mask2Former和BinsFormer用于3D目标检测、实例分割和深度估计任务。最终提交的是一个采用InternImage-L骨干网络的单一模型,在SHIFT验证集上取得了49.6的总分(29.5 Det mAP、80.3 mTPS、46.4 Seg mAP和7.93 silog)。此外,我们在实验中提供了一些有趣的观察结果,这些观察可能有助于推动稠密视觉预测中多任务学习的发展。
英文摘要:
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.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。