【VALSE论文速览-96期】Label Matching Semi-Supervised Object Detection

发布人

论文题目：Label Matching Semi-Supervised Object Detection
作者列表：陈斌斌 (海康威视研究院)，陈伟杰(海康威视研究院，浙江大学)，杨世才(海康威视研究院)，禤韵怡(海康威视研究院)，宋杰(浙江大学)，谢迪(海康威视研究院)，浦世亮(海康威视研究院)，宋明黎(浙江大学)，庄越挺(浙江大学)
论文摘要：
Semi-supervised object detection has made significant progress with the development of mean teacher driven self-training. Despite the promising results, the label mismatch problem is not yet fully explored in the previous works, leading to severe confirmation bias during self-training. In this paper, we delve into this problem and propose a simple yet effective LabelMatch framework from two different yet complementary perspectives, i.e., distribution-level and instance-level. For the former one, it is reasonable to approximate the class distribution of the unlabeled data from that of the labeled data according to Monte Carlo Sampling. Guided by this weakly supervision cue, we introduce a re-distribution mean teacher, which leverages adaptive label-distribution-aware confidence thresholds to generate unbiased pseudo labels to drive student learning. For the latter one, there exists an overlooked label assignment ambiguity problem across teacher-student models. To remedy this issue, we present a novel label assignment mechanism for self-training framework, namely proposal self-assignment, which injects the proposals from student into teacher and generates accurate pseudo labels to match each proposal in the student model accordingly. Experiments on both MS-COCO and PASCAL-VOC datasets demonstrate the considerable superiority of our proposed framework to other state-of-the-arts.
论文信息：
[1] Binbin Chen, Weijie Chen, Shicai Yang, et al. Label Matching Semi-Supervised Object Detection. CVPR 2022.
论文链接：
Label Matching Semi-Supervised Object Detection (thecvf.com)
Chen_Label_Matching_Semi-Supervised_CVPR_2022_supplemental.pdf (thecvf.com)
代码链接：
https://github.com/hikvision-research/SSOD
视频讲者简介：
陈斌斌，海康威视研究院算法工程师，毕业于华中科技大学人工智能学院。主要研究方向为目标检测，半监督目标检测和多模态等。

打开封面下载高清视频观看高清视频视频下载器

【VALSE论文速览-96期】Label Matching Semi-Supervised Object Detection

【VALSE论文速览-63期】CrossDet: Crossline Representation for Object Detection

【VALSE论文速览-02期】探索简单孪生网络表征学习

【VALSE论文速览-94期】Learning Domain Adaptive Object Detection with Probabilistic……

【VALSE论文速览-112期】Flow-Guided Transformer for Video Inpainting

【VALSE论文速览-159期】Dynamic Snake Convolution based on Topological Geometric ……

【VALSE论文速览-58期】Background-Click Supervision for Temporal Action Localization

【VALSE论文速览-57期】DeepMIH: Deep Invertible Network for Multiple Image Hiding

【VALSE论文速览-51期】Cross-Encoder for Unsupervised Gaze Representation Learning

【VALSE论文速览-64期】用于图像盲超分辨的无监督退化表示学习

【VALSE论文速览-149期】Robust Test-Time Adaptation in Dynamic Scenarios

【VALSE论文速览-30期】基于并行解码的端到端密集视频描述

【VALSE论文速览-109期】基于统一最优运输框架的通用域适应方法

【VALSE论文速览-98期】基于多面注意力机制的人群计数

20200731-Valse Student Seminar 谢雨彤《A Mutual Bootstrapping Model for Automated……》

【VALSE论文速览-123期】A Closer Look at Few-shot Classification Again

【VALSE论文速览-68期】MixFormer:更加简洁的端到端单目标跟踪器

【VALSE论文速览-114期】An Investigation into Whitening Loss for Self-supervised Learnin

【VALSE论文速览-81期】Deep Rectangling for Image Stitching: A Learning Baseline

【VALSE论文速览-01期】NeuralRecon: 基于单目视频的实时三维场景重建

【VALSE论文速览-76期】Robust Region Feature Synthesizer for Zero-Shot Object Detection

【VALSE论文速览-87期】Not Just Selection, but Exploration: 双视角一致的在线增量学习

【VALSE论文速览-52期】Class-aware Sounding Objects Localization via Audiovisual Corresp

【VALSE论文速览-77期】Semantic-Aware Domain Generalized Segmentation

【VALSE论文速览-88期】MSG-Transformer: Exchanging Local Spatial Information by……

20220123 VALSE Student Webinar【坐而论道or绝地反击：如何进行一次成功的Rebuttal？】李一鸣：从审稿人的视角浅析如何撰写……

【VALSE论文速览-50期】Trash or Treasure? An Interactive Dual-Stream Strategy for Single

【VALSE论文速览-124期】Base and Meta: A New Perspective on Few-shot Segmentation

【VALSE论文速览-97期】Adversarial Sticker: A Stealthy Attack Method in the Physical……

【VALSE论文速览-22期】基于噪声鲁棒对比损失的部分视图对齐表示学习

【VALSE论文速览-17期】基于对比预测的缺失视图聚类方法

20240717【面向事件相机的物体检测与跟踪】王逍：Visual Object Tracking using an Event Camera

【VALSE2024】0505 赵恒爽《APR：视觉基础大模型》

【VALSE论文速览-26期】面向视频目标跟踪的对抗模糊攻击

【VALSE论文速览-72期】MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose……

【VALSE论文速览-35期】Oriented R-CNN: 有向目标检测

【VALSE论文速览-172期】Learning to Upsample by Learning to Sample

【VALSE论文速览-198期】：AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ...

【VALSE论文速览-74期】Learning What Not to Segment: A New Perspective on Few-Shot……

【VALSE论文速览-11期】GeoSim: 自动驾驶中高清逼真的相机仿真

【VALSE论文速览-20期】GAIA：一个目标检测下游迁移框架