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

ISBN-13: 9783031197772 / Angielski / Miękka / 2022 / 749 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

AU-Aware 3D Face Reconstruction through Personalized AU-Specific Blendshape Learning.- B´ezierPalm: A Free Lunch for Palmprint Recognition.- Adaptive Transformers for Robust Few-Shot Cross-Domain Face Anti-Spoofing.- Face2Faceρ: Real-Time High-Resolution One-Shot Face Reenactment.- Towards Racially Unbiased Skin Tone Estimation via Scene Disambiguation.- BoundaryFace: A Mining Framework with Noise Label Self-Correction for Face Recognition.- Pre-training Strategies and Datasets for Facial Representation Learning.- Look Both Ways: Self-Supervising Driver Gaze Estimation and Road Scene Saliency.- MFIM: Megapixel Facial Identity Manipulation.- 3D Face Reconstruction with Dense Landmarks.- Emotion-Aware Multi-View Contrastive Learning for Facial Emotion Recognition.- Order Learning Using Partially Ordered Data via Chainization.- Unsupervised High-Fidelity Facial Texture Generation and Reconstruction.- Multi-Domain Learning for Updating Face Anti-Spoofing Models.- Towards Metrical Reconstruction of Human Faces.- Discover and Mitigate Unknown Biases with Debiasing Alternate Networks.- Unsupervised and Semi-Supervised Bias Benchmarking in Face Recognition.- Towards Efficient Adversarial Training on Vision Transformers.- MIME: Minority Inclusion for Majority Group Enhancement of AI Performance.- Studying Bias in GANs through the Lens of Race.- Trust, but Verify: Using Self-Supervised Probing to Improve Trustworthiness.- Learning to Censor by Noisy Sampling.- An Invisible Black-Box Backdoor Attack through Frequency Domain.- FairGRAPE: Fairness-Aware GRAdient Pruning mEthod for Face Attribute Classification.- Attaining Class-Level Forgetting in Pretrained Model Using Few Samples.- Anti-Neuron Watermarking: Protecting Personal Data against Unauthorized Neural Networks.- An Impartial Take to the CNN vs Transformer Robustness Contest.- Recover Fair Deep Classification Models via Altering Pre-trained Structure.- Decouple-and-Sample: Protecting Sensitive Information in Task

Agnostic Data Release.- Privacy-Preserving Action Recognition via Motion Difference Quantization.- Latent Space Smoothing for Individually Fair Representations.- Parameterized Temperature Scaling for Boosting the Expressive Power

in Post-Hoc Uncertainty Calibration.- FairStyle: Debiasing StyleGAN2 with Style Channel Manipulations.- Distilling the Undistillable: Learning from a Nasty Teacher.- SOS! Self-Supervised Learning over Sets of Handled Objects in Egocentric Action Recognition.- Egocentric Activity Recognition and Localization on a 3D Map.- Generative Adversarial Network for Future Hand Segmentation from Egocentric Video.- My View Is the Best View: Procedure Learning from Egocentric Videos.- GIMO: Gaze-Informed Human Motion Prediction in Context.- Image-Based CLIP-Guided Essence Transfer.- Detecting and Recovering Sequential DeepFake Manipulation.- Self-Supervised Sparse Representation for Video Anomaly Detection.