代码已开源在： https://github.com/TuSimple/SST 研究简介： 在自动驾驶场景中，相比于整个场景的尺度，单个物体的尺度通常很小。下图展示了COCO数据集和Waymo数据集上物体相对尺度的分布情况： COCO和Waymo上物体相对尺度分布 这一特性往往被基于Pillar或者体素的检测器所忽略，它们通常借用了成熟的2D多尺度检测器的结构。基于这一考量，本文探索了单步长（无降采样）的检测器结构。如果简单地将卷积网络提升为单步长网络，会取得一定的性能提升，但是会带来感受野不足的问题以及巨大的计算量。为了得到一个高效高性能的单步长检测器，我们借用了当前流行的swin transformer的结构，舍弃了其多尺度的结构并且针对点云数据的特点将其稀疏化，我们将其命名为单步长稀疏Transformer（Single-stride Sparse Transformer, SST）。我们在当前最大的3D检测数据集Waymo Open Dataset上做了详尽的实验，从各个方面探讨了SST的特性，并取得了SoTA的性能，特别是在小物体上比之前的方法有了显著的提升（达到了83.8的Level 1 AP）。 Sparse Attention结构设计
Research
Instance Guided Proposal Network for Person Search (CVPR2020, oral)
Journal / Conference Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 2020) [PDF link: link] [Code link: link] Keywords Person Search, Siamese RPN, Relation Block Abstract […]
Bi-directional Interaction Network for Person Search (CVPR2020)
Journal / Conference Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 2020) [PDF link: link] [Code link: link] Keywords Person Search, Bi-directional Interaction Abstract Existing works […]
Learning Integral Objects with Intra-Class Discriminator for Weakly-Supervised Semantic Segmentation (CVPR2020)
Journal / Conference Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 2020) [PDF link: PENDING] [Code link: here] Keywords Weakly-Supervised Learning, Semantic Segmentation Abstract Image-level weakly-supervised […]
CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation (AAAI2019)
Journal / Conference Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI, 2020) [PDF link: here] [Code link: here] Keywords Weakly-Supervised Learning, Semantic Segmentation Abstract Weakly supervised semantic segmentation with only image-level […]
Efficient Neural Architecture Transformation Search in Channel-Level for Object Detection (NeurIPS2019)
Journal / Conference Thirty-third Conference on Neural Information Processing Systems(NIPS, 2019) [PDF link: here] [Code link: PENDING] Keywords Neural Architecture Transformation Search(NATS), Object Detection Abstract Recently, Neural Architecture Search has […]
POD: Practical Object Detection with Scale-Sensitive Network (ICCV2019)
Journal / Conference The IEEE International Conference on Computer Vision (ICCV, 2019) [PDF link: here] [Code link: Pending] Keywords Scale-sensitive object detection,Global Scale Learning module Abstract Scale-sensitive object detection remains […]
Sequence Level Semantics Aggregation for Video Object Detection (ICCV2019)
Journal / Conference The IEEE International Conference on Computer Vision (ICCV, 2019) [PDF link:here] [Code link: here] Keywords Video Object Detection, SEquence Level Semantics Aggregation (SELSA) Abstract Video objection detection […]
Scale-Aware Trident Networks for Object Detection (ICCV2019)
Journal / Conference The IEEE International Conference on Computer Vision (ICCV, 2019) [PDF link: here] [Code link: here] Keywords Object Detection, TridentNet Abstract Scale variation is one of the key […]
Spectral Feature Transformation for Person Re-Identification (ICCV2019)
Journal / Conference IEEE International Conference on Computer Vision (ICCV2019) [PDF link: here] [Code link: here] Keywords Person Re-ID, Spectral Feature Transformation, Post-processing Abstract With the surge of deep learning […]
