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

ISBN-13: 9783031197963 / Angielski / Miękka / 2022 / 757 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

Dynamic Dual Trainable Bounds for Ultra-Low Precision Super-Resolution Networks.- OSFormer: One-Stage Camouflaged Instance Segmentation with Transformers.- Highly Accurate Dichotomous Image Segmentation.- Boosting Supervised Dehazing Methods via Bi-Level Patch Reweighting.- Flow-Guided Transformer for Video Inpainting.- Shift-tolerant Perceptual Similarity Metric.- Perception-Distortion Balanced ADMM Optimization for Single-Image Super-Resolution.- VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder.- Uncertainty Learning in Kernel Estimation for Multi-stage Blind Image Super-Resolution.- Learning Spatio-Temporal Downsampling for Effective Video Upscaling.- Learning Local Implicit Fourier Representation for Image Warping.- SepLUT: Separable Image-Adaptive Lookup Tables for Real-Time Image Enhancement.- Blind Image Decomposition.- MuLUT: Cooperating Multiple Look-Up Tables for Efficient Image Super-Resolution.- Learning Spatiotemporal Frequency-Transformer for Compressed Video Super-Resolution.- Spatial-Frequency Domain Information Integration for Pan-Sharpening.- Adaptive Patch Exiting for Scalable Single Image Super-Resolution.- Efficient Meta-Tuning for Content-Aware Neural Video Delivery.- Reference-Based Image Super-Resolution with Deformable Attention Transformer.- Local Color Distributions Prior for Image Enhancement.- L-CoDer: Language-Based Colorization with Color-Object Decoupling Transformer.- From Face to Natural Image: Learning Real Degradation for Blind Image Super-Resolution.- Towards Interpretable Video Super-Resolution via Alternating Optimization.- Event-Based Fusion for Motion Deblurring with Cross-Modal Attention.- Fast and High Quality Image Denoising via Malleable Convolution.- TAPE: Task-Agnostic Prior Embedding for Image Restoration.- Uncertainty Inspired Underwater Image Enhancement.- Hourglass Attention Network for Image Inpainting.- Unfolded Deep Kernel Estimation for Blind Image Super-Resolution.- Event-Guided Deblurring of Unknown Exposure Time Videos.- ReCoNet: Recurrent Correction Network for Fast and Efficient Multi-Modality Image Fusion.- Content Adaptive Latents and Decoder for Neural Image Compression.- Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution.- Unidirectional Video Denoising by Mimicking Backward Recurrent Modules with Look-Ahead Forward Ones.- Self-Supervised Learning for Real-World Super-Resolution from Dual Zoomed Observations.- Secrets of Event-Based Optical Flow.- Towards Efficient and Scale-Robust Ultra-High-Definition Image Demoir´eing.- ERDN: Equivalent Receptive Field Deformable Network for Video Deblurring.- Rethinking Generic Camera Models for Deep Single Image Camera Calibration to Recover Rotation and Fisheye Distortion.- ART-SS: An Adaptive Rejection Technique for Semi-Supervised Restoration for Adverse Weather-Affected Images.- Fusion from Decomposition: A Self-Supervised Decomposition Approach for Image Fusion.-Learning Degradation Representations for Image Deblurring.