arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2204.04656cs.CV
收藏

视频K-Net:一个简单,强大,统一的视频分割基线

Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation

Peking University(北京大学);Nanyang Technological University(南洋理工大学);SenseTime Research(商汤科技研究院);The Chinese University of Hong Kong(香港中文大学);Shanghai AI Lab(上海人工智能实验室)

Xiangtai Li, Wenwei Zhang, Jiangmiao Pang, Kai Chen, Guangliang Cheng, Yunhai Tong, Chen Change Loy

摘要:

本文介绍了视频K-Net,一个简单,强大,统一的框架,完全端到端的视频全景分割。该方法建立在K-Net之上,K-Net是一种通过一组可学习的内核统一图像分割的方法。我们观察到,这些来自K-Net的可学习内核(对对象外观和上下文进行编码)可以自然地将视频帧中的相同实例关联起来。受到这一观察的启发,视频K-Net学会了通过简单的基于内核的外观建模和跨时间内核交互来同时分割和跟踪视频中的“事物”和“东西”。尽管简单,它实现了国家的最先进的视频全景分割的结果城市景观VPS,KITTI-STEP和VIPSeg没有花里胡哨。特别是在KITTI-STEP上,该方法比以往的方法提高了近12%的相对性能。在VIPSeg上,视频K-Net提高了近15%的相对改进,并导致39.8%的VPQ。我们还验证了它的泛化视频语义分割,我们提高了2%的VSPW数据集上的各种基线。此外,我们将K-Net扩展到剪辑级视频框架中进行视频实例分割,其中我们在YouTube-2019验证集上获得了ResNet 50骨干的40.5% mAP和Swin-base的54.1% mAP。我们希望这种简单而有效的方法可以作为统一的视频分割设计的一个新的,灵活的基线。代码和模型都在https://github.com/lxtGH/Video-K-Net上发布。

英文摘要:

This paper presents Video K-Net, a simple, strong, and unified framework for fully end-to-end video panoptic segmentation. The method is built upon K-Net, a method that unifies image segmentation via a group of learnable kernels. We observe that these learnable kernels from K-Net, which encode object appearances and contexts, can naturally associate identical instances across video frames. Motivated by this observation, Video K-Net learns to simultaneously segment and track "things" and "stuff" in a video with simple kernel-based appearance modeling and cross-temporal kernel interaction. Despite the simplicity, it achieves state-of-the-art video panoptic segmentation results on Citscapes-VPS, KITTI-STEP, and VIPSeg without bells and whistles. In particular, on KITTI-STEP, the simple method can boost almost 12\% relative improvements over previous methods. On VIPSeg, Video K-Net boosts almost 15\% relative improvements and results in 39.8 % VPQ. We also validate its generalization on video semantic segmentation, where we boost various baselines by 2\% on the VSPW dataset. Moreover, we extend K-Net into clip-level video framework for video instance segmentation, where we obtain 40.5% mAP for ResNet50 backbone and 54.1% mAP for Swin-base on YouTube-2019 validation set. We hope this simple, yet effective method can serve as a new, flexible baseline in unified video segmentation design. Both code and models are released at https://github.com/lxtGH/Video-K-Net.

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。

↑