Part 1: Learning Optical Expansion from Scale Matching ---- Towards All-in-one Pre-training via Maximizing Multi-modal Mutual Informationn
Part 2: CORA: Adapting CLIP for Open-Vocabulary Detection with Region Prompting and Anchor Pre-Matching ---- Are We Ready for Vision-Centric Driving Streaming Perception? The ASAP Benchmark
Part 3: Vision Transformers are Good Mask Auto-Labelers ---- ECON: Explicit Clothed humans Optimized via Normal integration
Part 4: Zero-shot Generative Model Adaptation via Image-specific Prompt Learning ---- ALOFT: A Lightweight MLP-like Architecture with Dynamic Low-frequency Transform for Domain Generalization
Part 5: Token Boosting for Robust Self-Supervised Visual Transformer Pre-training ---- Mask-guided Matting in the Wild
- DA Wand: Distortion-Aware Selection using Neural Mesh Parameterization
- NeuralEditor: Editing Neural Radiance Fields via Manipulating Point Clouds
- EditableNeRF: Editing Topologically Varying Neural Radiance Fields by Key Points
- JacobiNeRF: NeRF Shaping with Mutual Information Gradients
- NeuFace: Realistic 3D Neural Face Rendering from Multi-view Images
- NeFII: Inverse Rendering for Reflectance Decomposition with Near-Field Indirect Illumination
- NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects
- RelightableHands: Efficient Neural Relighting of Articulated Hand Models
- Multi-View Azimuth Stereo via Tangent Space Consistency
- VDN-NeRF: Resolving Shape-Radiance Ambiguity via View-Dependence Normalization
- ReLight My NeRF: A Dataset for Novel View Synthesis and Relighting of Real World Objects
- Looking Through the Glass: Neural Surface Reconstruction Against High Specular Reflections
- HandNeRF: Neural Radiance Fields for Animatable Interacting Hands
- Generalizable Implicit Neural Representations via Instance Pattern Composers
- Semi-supervised Hand Appearance Recovery via Structure Disentanglement and Dual Adversarial Discrimination
- NeRFLix: High-Quality Neural View Synthesis by Learning a Degradation-Driven Inter-Viewpoint MiXer
- CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition
- DP-NeRF: Deblurred Neural Radiance Field with Physical Scene Priors
- NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects
- HairStep: Transfer Synthetic to Real Using Strand and Depth Maps for Single-View 3D Hair Modeling
- GM-NeRF: Learning Generalizable Model-Based Neural Radiance Fields From Multi-View Images
- Learning Visibility Field for Detailed 3D Human Reconstruction and Relighting
- ReLight My NeRF: A Dataset for Novel View Synthesis and Relighting of Real World Objects
- MAIR: Multi-view Attention Inverse Rendering with 3D Spatially-Varying Lighting Estimation
- Relightable Neural Human Assets from Multi-view Gradient Illumination
- InstantAvatar: Learning Avatars from Monocular Video in 60 Seconds
- Compressing Volumetric Radiance Fields to 1 MB
- Neuralangelo: High-Fidelity Neural Surface Reconstruction
- NeRFLight: Fast and Light Neural Radiance Fields using a Shared Feature Grid
- Real-Time Neural Light Field on Mobile Devices
- F2-NeRF: Fast Neural Radiance Field Training With Free Camera Trajectories
- Latent-NeRF for Shape-Guided Generation of 3D Shapes and Textures
- OmniAvatar: Geometry-Guided Controllable 3D Head Synthesis
- Diffusion-Based Signed Distance Fields for 3D Shape Generation
- DiffusioNeRF: Regularizing Neural Radiance Fields With Denoising Diffusion Models
- MonoAvatar: Learning Personalized High Quality Volumetric Head Avatars from Monocular RGB Videos
- Learning Neural Volumetric Representations of Dynamic Humans in Minutes
- Structural Multiplane Image: Bridging Neural View Synthesis and 3D Reconstruction
- DINER: Depth-aware Image-based NEural Radiance fields
- FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularization
- MixNeRF: Modeling a Ray with Mixture Density for Novel View Synthesis from Sparse Inputs
- BlendFields: Few-Shot Example-Driven Facial Modeling
- Towards Unbiased Volume Rendering of Neural Implicit Surfaces with Geometry Priors
- gSDF: Geometry-Driven Signed Distance Functions for 3D Hand-Object Reconstruction
- DiffusioNeRF: Regularizing Neural Radiance Fields With Denoising Diffusion Models
- ECON: Explicit Clothed humans Optimized via Normal integration
- CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition
- Regularize Implicit Neural Representation by Itself
- NeUDF: Leaning Neural Unsigned Distance Fields With Volume Rendering
- NeuralUDF: Learning Unsigned Distance Fields for Multi-View Reconstruction of Surfaces With Arbitrary Topologies
- NeAT: Learning Neural Implicit Surfaces with Arbitrary Topologies from Multi-view Images
- ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision
- NEF: Neural Edge Fields for 3D Parametric Curve Reconstruction from Multi-view Images