Computer Vision - ECCV 2022

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 2327, 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.

Inhalt
Learning Depth from Focus in the Wild.- Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World.- An End-to-End Transformer Model for Crowd Localization.- Few-Shot Single-View 3D Reconstruction with Memory Prior Contrastive Network.- DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection.- Adaptive Co-Teaching for Unsupervised Monocular Depth Estimation.- Fusing Local Similarities for Retrieval-Based 3D Orientation Estimation of Unseen Objects.- Lidar Point Cloud Guided Monocular 3D Object Detection.- Structural Causal 3D Reconstruction.- 3D Human Pose Estimation Using Mbius Graph Convolutional Networks.- Learning to Train a Point Cloud Reconstruction Network without Matching.- PanoFormer: Panorama Transformer for Indoor 360° Depth Estimation.- Self-supervised Human Mesh Recovery with Cross-Representation Alignment.- AlignSDF: Pose-Aligned Signed Distance Fields for Hand-Object Reconstruction.- A Reliable Online Method for Joint Estimation of Focal Length and Camera Rotation.- PS-NeRF: Neural Inverse Rendering for Multi-View Photometric Stereo.- Share with Thy Neighbors: Single-View Reconstruction by Cross-Instance Consistency.- Towards Comprehensive Representation Enhancement in Semantics- Guided Self-Supervised Monocular Depth Estimation.- AvatarCap: Animatable Avatar Conditioned Monocular Human Volumetric Capture.- Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers.- GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping.- Multi-modal Masked Pre-training for Monocular Panoramic Depth Completion.- GitNet: Geometric Prior-Based Transformation for Birds-Eye View Segmentation.- Learning Visibility for Robust Dense Human Body Estimation.- Towards High-Fidelity Single-View Holistic Reconstructionof Indoor Scenes.- CompNVS: Novel View Synthesis with Scene Completion.- SketchSampler: Sketch-Based 3D Reconstruction via View-Dependent Depth Sampling.- LocalBins: Improving Depth Estimation by Learning Local Distributions.- 2D GANs Meet Unsupervised Single-View 3D Reconstruction.- InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images.- Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors.- Bilateral Normal Integration.- S2Contact: Graph-Based Network for 3D Hand-Object Contact Estimation with Semi-Supervised Learning.- SC-wLS: Towards Interpretable Feed-Forward Camera Re-localization.- FloatingFusion: Depth from ToF and Image-Stabilized Stereo Cameras.- DELTAR: Depth Estimation from a Light-Weight ToF Sensor and RGB Image.- 3D Room Layout Estimation from a Cubemap of Panorama Image via Deep Manhattan Hough Transform.- RBP-Pose: ResidualBounding Box Projection for Category-Level Pose Estimation.- Monocular 3D Object Reconstruction with GAN Inversion.- Map-Free Visual Relocalization: Metric Pose Relative to a Single Image.- Self-Distilled Feature Aggregation for Self-Supervised Monocular Depth Estimation.- Planes vs. Chairs: Category-Guided 3D Shape Learning without Any 3D Cues.