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

Editing Out-of-Domain GAN Inversion via Differential Activations.- On the Robustness of Quality Measures for GANs.- Sound-Guided Semantic Video Generation.- Inpainting at Modern Camera Resolution by Guided PatchMatch with Auto-Curation.- Controllable Video Generation through Global and Local Motion Dynamics.- StyleHEAT: One-Shot High-Resolution Editable Talking Face Generation via Pre-trained StyleGAN.- Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer.- Combining Internal and External Constraints for Unrolling Shutter in Videos.- WISE: Whitebox Image Stylization by Example-Based Learning.- Neural Radiance Transfer Fields for Relightable Novel-View Synthesis with Global Illumination.- Transformers As Meta-Learners for Implicit Neural Representations.- Style Your Hair: Latent Optimization for Pose-Invariant Hairstyle Transfer via Local-Style-Aware Hair Alignment.- High-Resolution Virtual Try-On withMisalignment and OcclusionHandled Conditions.- A Codec Information Assisted Framework for Efficient Compressed Video Super-Resolution.- Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait Image Synthesis.- AdaNeRF: Adaptive Sampling for Real-Time Rendering of Neural Radiance Fields.- Improving the Perceptual Quality of 2D Animation Interpolation.- Selective TransHDR: Transformer-Based Selective HDR Imaging Using Ghost Region Mask.- Learning Series-Parallel Lookup Tables for Efficient Image Super-Resolution.- GeoAug: Data Augmentation for Few-Shot NeRF with Geometry Constraints.- DoodleFormer: Creative Sketch Drawing with Transformers.- Implicit Neural Representations for Variable Length Human Motion Generation.- Learning Object Placement via Dual-Path Graph Completion.- Expanded Adaptive Scaling Normalization for End to End Image Compression.- Generator Knows What Discriminator Should Learn inUnconditional GANs.- Compositional Visual Generation with Composable Diffusion Models.- ManiFest: Manifold Deformation for Few-Shot Image Translation.- ManiFest: Manifold Deformation for Few-Shot Image Translation.- Supervised Attribute Information Removal and Reconstruction for Image Manipulation.- BLT: Bidirectional Layout Transformer for Controllable Layout Generation.- Diverse Generation from a Single Video Made Possible.- Rayleigh EigenDirections (REDs): Nonlinear GAN Latent Space Traversals for Multidimensional Features.- Bridging the Domain Gap towards Generalization in Automatic Colorization.- Generating Natural Images with Direct Patch Distributions Matching.- Context-Consistent Semantic Image Editing with Style-Preserved Modulation.- Eliminating Gradient Conflict in Reference-Based Line-Art Colorization.- Unsupervised Learning of Efficient Geometry-Aware Neural Articulated Representations.- JPEG Artifacts Removal viaContrastive Representation Learning.- Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning.- Efficient Long-Range Attention Network for Image Super-Resolution.- FlowFormer: A Transformer Architecture for Optical Flow.- Coarse-to-Fine Sparse Transformer for Hyperspectral Image Reconstruction.- Learning Shadow Correspondence for Video Shadow Detection.- Metric Learning Based Interactive Modulation for Real-World Super-Resolution.