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

StyleGAN-Human.A Data-Centric Odyssey of Human Generation.- ColorFormer. Image Colorization via Color Memory Assisted Hybrid-Attention Transformer.- EAGAN. Efficient Two-Stage Evolutionary Architecture Search for GANs.- Weakly-Supervised Stitching Network for Real-World Panoramic Image Generation.- DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation.- Multimodal Conditional Image Synthesis with Product-of-Experts GANs.- Auto-Regressive Image Synthesis with Integrated Quantization.- JoJoGAN. One Shot Face Stylization.- VecGAN: Image-to-Image Translation with Interpretable Latent Directions.- Any-Resolution Training for High-Resolution Image Synthesis.- CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer.- CANF-VC: Conditional Augmented Normalizing Flows for Video Compression.- Bi-Level Feature Alignment for Versatile Image Translation and Manipulation.- High-Fidelity Image Inpainting with GAN Inversion.- DeltaGAN: Towards Diverse Few-Shot Image Generation with Sample-Specific Delta.- Image Inpainting with Cascaded Modulation GAN and Object-Aware Training.- StyleFace: Towards Identity-Disentangled Face Generation on Megapixels.- Video Extrapolation in Space and Time.- Contrastive Learning for Diverse Disentangled Foreground Generation.- BIPS.Bi-modal Indoor Panorama Synthesis via Residual Depth-Aided Adversarial Learning.- Augmentation of rPPG Benchmark Datasets: Learning to Remove and Embed rPPG Signals via Double Cycle Consistent Learning from Unpaired Facial Videos.- Geometry-Aware Single-Image Full-Body Human Relighting.- 3D-Aware Indoor Scene Synthesis with Depth Priors.- Deep Portrait Delighting.- Vector Quantized Image-to-Image Translation.- The Surprisingly Straightforward Scene Text Removal Method with Gated Attention and Region of Interest Generation: A Comprehensive Prominent Model Analysis.- Free-Viewpoint RGB-D Human Performance Capture and Rendering.- Multiview Regenerative Morphing with Dual Flows.- Hallucinating Pose-Compatible Scenes.- Motion and Appearance Adaptation for Cross-Domain Motion Transfer.- Layered Controllable Video Generation.- Custom Structure Preservation in Face Aging.- Spatio-Temporal Deformable Attention Network for Video Deblurring.- NeuMesh: Learning Disentangled Neural Mesh-Based Implicit Field for Geometry and Texture Editing.- NeRF for Outdoor Scene Relighting.- CoGS: Controllable Generation and Search from Sketch and Style.- HairNet: Hairstyle Transfer with Pose Changes.- Unbiased Multi-Modality Guidance for Image Inpainting.- Intelli-Paint: Towards Developing More Human-Intelligible Painting Agents.- Motion Transformer for Unsupervised Image Animation.- NUWA: Visual Synthesis Pre-training for Neural visUal World creAtion.- EleGANt: Exquisite and Locally Editable GAN forMakeup Transfer.