In this paper, we present a joint end-to-end line segment detection algorithm using Transformers that is post-processing and heuristics-guided intermediate processing (edge/junction/region detection) free. Our method, named LinE segment TRansformers (LETR), takes advantages of having integrated tokenized queries, a self-attention mechanism, and an encoding-decoding strategy within Transformers by skipping standard heuristic designs for the edge element detection and perceptual grouping processes. We equip Transformers with a multi-scale encoder/decoder strategy to perform fine-grained line segment detection under a direct endpoint distance loss. This loss term is particularly suitable for detecting geometric structures such as line segments that are not conveniently represented by the standard bounding box representations. The Transformers learn to gradually refine line segments through layers of self-attention. In our experiments, we show state-of-the-art results on Wireframe and YorkUrban benchmarks.
TraSeTR: Track-to-Segment Transformer with Contrastive Query for Instance-level Instrument Segmentation in Robotic Surgery
The Chinese University of Hong Kong（香港中文大学）；Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；University College London（伦敦大学学院）
Spatio-Temporal Multi-Task Learning Transformer for Joint Moving Object Detection and Segmentation
Valeo R&D（法雷奥研发中心）
移动物体对于自动驾驶任务具有特殊的重要性。检测移动对象可以通过分割对象像素而被定位为移动对象分割，或者通过生成移动目标的限界框而被定位为移动目标检测。在本文中，我们提出了一个多任务学习架构，基于Transformers，通过一个网络来共同执行这两个任务。由于运动特征对任务的重要性，整个设置基于时空聚合。我们评估的性能与MTL设置，无论是早期共享的编码器，和后期共享的编码器-解码器Transformers器的单独任务架构。对于后者，我们提出了一种新的联合任务查询解码器Transformer，使我们能够有任务专用头的共享模型。为了评估我们的方法，我们使用KITTI MOD [29]数据集。结果显示，在单个任务网络上，移动目标检测的mAP提高了1.5%，移动对象分割的IoU提高了2%。
