推荐理事:林宙辰原文标题:VIL-100: A New Dataset and A Baseline Model for Video Instance Lane Detection原文链接:https://arxiv.org/abs/2108.08482原文代码链接: https://github.com/yujun0-0/MMA-Net数据集链接:https://pan.baidu.com/s/1NkP_5LMLTn6qsu9pSbyi0g 提取码:iy16
图 6:可视化比较 【消融实验】表3中第一列(“Basic”)是从MMA-Net中去除局部全局记忆和多尺度聚合机制。第二列(“+LM”)是将局部记忆添加到“Basic”,而第三列(“+GM”)是将全局记忆添加到“Basic”。第四列(“+LGM”)是将局部和全局记忆均添加到“Basic”, 也即从我们的网络中移除多尺度聚合机制。消融实验充分证明了局部和全局记忆模块以及多尺度聚合模式的有效性。表 3:消融实验定量评估
参考文献
[1] Davy Neven, Bert De Brabandere, Stamatios Georgoulis, Marc Proesmans, and Luc Van Gool. Towards end-to-end lane detection: an instance segmentation approach. In IVS, pages 286–291, 2018.[2] Xingang Pan, Jianping Shi, Ping Luo, XiaogangWang, and Xiaoou Tang. Spatial as deep: Spatial cnn for traffific sceneunderstanding. In AAAI, pages 7276–7283, 2018.[3] Yuenan Hou, Zheng Ma, Chunxiao Liu, and ChenChange Loy. Learning lightweight lane detection cnns by self attentiondistillation. In ICCV, pages 1013–1021, 2019.[4] Zequn Qin, Huanyu Wang, and Xi Li. Ultra fast structureaware deep lane detection. In ECCV, pages 276–291. Springer, 2020.[5]Ruijin Liu, Zejian Yuan, Tie Liu, and Zhiliang Xiong. End-to-end lane shape prediction with transformers. In WACV, pages 3694–3702, 2021.
[6] Joakim Johnander, Martin Danelljan, Emil Brissman, Fahad Shahbaz Khan, and Michael Felsberg. A generative appearance model for end-to-end video object segmentation. In CVPR, pages 8953–8962, 2019.
[7] Carles Ventura, Miriam Bellver, Andreu Girbau, Amaia Salvador, Ferran Marques, and Xavier Giro i Nieto. RVOS: Endto-end recurrent network for video object segmentation. In CVPR, pages 5277–5286, 2019.[8] Seoung Wug Oh, Joon-Young Lee, Ning Xu, and Seon Joo Kim. Video object segmentation using space-time memory networks. In ICCV, pages 9226–9235, 2019.[9] Yongqing Liang, Xin Li, Navid Jafari, and Qin Chen. Video object segmentation with adaptive feature bank and uncertain-region refifinement. In NeurIPS, 2020.[10] Yizhuo Zhang, Zhirong Wu, Houwen Peng, and Stephen Lin. A transductive approach for video object segmentation. In CVPR, pages 6949–6958, 2020.