Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part X

ISBN-13: 9783031200793 / Angielski / Miękka / 2022 / 759 str.

Shai Avidan; Gabriel Brostow; Moustapha Cissé

Opis

The 39-volume set, comprising the LNCS books 13661 until 13699, constitutes the refereed proceedings of the 17th European Conference on Computer Vision, ECCV 2022, held in Tel Aviv, Israel, during October 23–27, 2022. The 1645 papers presented in these proceedings were carefully reviewed and selected from a total of 5804 submissions. The papers deal with topics such as computer vision; machine learning; deep neural networks; reinforcement learning; object recognition; image classification; image processing; object detection; semantic segmentation; human pose estimation; 3d reconstruction; stereo vision; computational photography; neural networks; image coding; image reconstruction; object recognition; motion estimation.

Spis treści

DFNet: Enhance Absolute Pose Regression with Direct Feature Matching.- Cornerformer: Purifying Instances for Corner-Based Detectors.- Robust Object Detection with Inaccurate Bounding Boxes.- Efficient Decoder-Free Object Detection with Transformers.- Cross-Modality Knowledge Distillation Network for Monocular 3D Object Detection.- ReAct: Temporal Action Detection with Relational Queries.- Towards Accurate Active Camera Localization.- Camera Pose Auto-Encoders for Improving Pose Regression.- Improving the Intra-Class Long-Tail in 3D Detection via Rare Example Mining.- Bagging Regional Classification Activation Maps for Weakly Supervised Object Localization.- UC-OWOD: Unknown-Classified Open World Object Detection.- RayTran: 3D Pose Estimation and Shape Reconstruction of Multiple

Objects from Videos with Ray-Traced Transformers.- GTCaR: Graph Transformer for Camera Re-Localization.- 3D Object Detection with a Self-Supervised Lidar Scene Flow Backbone.- Open Vocabulary Object Detection with Pseudo Bounding-Box Labels.- Few-Shot Object Detection by Knowledge Distillation Using Bag-of-Visual-Words Representations.- SALISA: Saliency-Based Input Sampling for Efficient Video Object

Detection.- ECO-TR: Efficient Correspondences Finding via Coarse-to-Fine

Refinement.- Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by

Radial Keypoint Voting.- Long-Tailed Instance Segmentation Using Gumbel Optimized Loss.- DetMatch: Two Teachers Are Better than One for Joint 2D and 3D

Semi-Supervised Object Detection.- ObjectBox: From Centers to Boxes for Anchor-Free Object Detection.- Is Geometry Enough for Matching in Visual Localization?.- SWFormer: Sparse Window Transformer for 3D Object Detection in Point Clouds.- PCR-CG: Point Cloud Registration via Deep Explicit Color and Geometry.- GLAMD: Global and Local Attention Mask Distillation for Object Detectors.- FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection.- Video Anomaly Detection by Solving Decoupled Spatio-Temporal

Jigsaw Puzzles.- Class-Agnostic Object Detection with Multi-modal Transformer.- Enhancing Multi-modal Features Using Local Self-Attention for 3D

Object Detection.- Object Detection As Probabilistic Set Prediction.- Weakly-Supervised Temporal Action Detection for Fine-Grained Videos with Hierarchical Atomic Actions.- Neural Correspondence Field for Object Pose Estimation.- On Label Granularity and Object Localization.- OIMNet++: Prototypical Normalization and Localization-Aware

Learning for Person Search.- Out-of-Distribution Identification: Let Detector Tell Which I Am Not Sure.- Learning with Free Object Segments for Long-Tailed Instance

Segmentation.- Autoregressive Uncertainty Modeling for 3D Bounding Box Prediction.- 3D Random Occlusion and Multi-layer Projection for Deep Multi-Camera Pedestrian Localization.- A Simple Single-Scale Vision Transformer for Object Detection and

Instance Segmentation.- Simple Open-Vocabulary Object Detection with Vision Transformers