CVPR 2026

3,659 papers

Top Research Directions

Research directions most represented by CVPR 2026 papers.

Generalizing from Synthetic Data to the Real World 54
Overcoming the Static Scene Assumption 47
Eliminating Per-Scene Optimization with Feed-Forward 3D Reconstruction 39
Scaling Visual Attention Beyond Quadratic Complexity 37
Mitigating Multi View Inconsistency in Generative 3D Modeling 35
The Spatial Reasoning Bottleneck in Multimodal Models 35
Aligning Global Semantics with Local Spatial Precision 33
The Decoupling of Visual Perception and Logical Reasoning 31
Aligning Video Generation Models with Physical Laws 29
Spatial Attention Masking in Generative Diffusion Models 29
Overcoming the Iterative Inference Latency of Diffusion Models 28
Physics Constrained Generative Human Motion Synthesis 26
Fusing 2D Semantic Features with 3D Geometric Representations 26
Group Relative Policy Optimization for Large Reasoning Models 25
Exploiting Temporal Redundancy in Video Neural Networks 24

Top Authors

AuthorPapers
Honglu Zhou4
Caiming Xiong3
Juan Carlos Niebles3
Junhao Chen3
Ramesh Raskar3
Siddharth Somasundaram3
silvio savarese3
Yanwei Fu3
Akshat Dave2
Andrew Zisserman2
Botong Wu2
Dahua Lin2
Di Wu2
Fangneng Zhan2
Gaowen Liu2
Gim Hee Lee2
Hanming Deng2
Hanspeter Pfister2
Hui Shan2
Jian Zhang2

Papers

Title
Rethinking BCE Loss for Multi-Label Image Recognition with Fine-Tuning
Rethinking Box Supervision: Bias-Free Weakly Supervised Medical Segmentation
Rethinking Camera Choice: An Empirical Study on Fisheye Camera Properties in Robotic Manipulation
Rethinking Concept Bottleneck Models: From Pitfalls to Solutions
Rethinking Cross-Modal Anchor Alignment for Mitigating Error Accumulation
Rethinking Dataset Distillation: Hard Truths about Soft Labels
Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features
Rethinking Intermediate Representation for VLM-based Robot Manipulation
Rethinking Knowledge Transfer in Image Quality Assessment: A Perceptual Preference Structure Alignment Perspective
Rethinking MLLM Itself as a Segmenter with a Single Segmentation Token
Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein Distance
Rethinking Pose Refinement in 3D Gaussian Splatting under Pose Prior and Geometric Uncertainty
Rethinking Position Embedding as a Context Controller for Multi-Reference and Multi-Shot Video Generation
Rethinking Prompt Design for Inference-time Scaling in Text-to-Visual Generation
Rethinking SNN Online Training and Deployment: Gradient-Coherent Learning via Hybrid-Driven LIF Model
Rethinking Token Reduction for Large Vision-Language Models
Rethinking Two-Stage Referring-by-Tracking in Referring Multi-Object Tracking: Make it Strong Again
Rethinking UMM Visual Generation: Masked Modeling for Efficient Image-Only Pre-training
Rethinking Visual Rearrangement from A Diffusion Perspective
RetimeGS: Continuous-Time Reconstruction of 4D Gaussian Splatting
RetouchIQ: MLLM Agents for Instruction-Based Image Retouching with Generalist Reward
Retrieve and Segment: Are a Few Examples Enough to Bridge the Supervision Gap in Open-Vocabulary Segmentation?
Retrieve-to-Restore: Efficient All-in-One Image Restoration with a Retrieval-Based Degradation Bank
Retrieving Counterfactuals Improves Visual In-Context Learning
RevINN: An End-to-End Invertible Neural Network for Reversible Adversarial Examples Generation
Revisiting 2D Foundation Models for Scalable 3D Medical Image Classification
Revisiting 3D Reconstruction Kernels as Low-Pass Filters
Revisiting F-measure Optimization in Multi-Label Classification: A Sampling-based Approach
Revisiting Learning with Noisy Labels: Active Forgetting and Noise Suppression
Revisiting Model Stitching In the Foundation Model Era
Revisiting Monocular SLAM with Spatio-Temporal Scene Modeling
Revisiting Multimodal KV Cache Compression: A Frequency-Domain-Guided Outlier-KV-Aware Approach
Revisiting Sparsity Constraint Under High-Rank Property in Partial Multi-Label Learning
Revisiting the Necessity of Full Accuracy: Weakly Supervised Object-Level Offset Correction for Misaligned Building Labels
Revisiting the Necessity of Lengthy Chain-of-Thought in Vision-centric Reasoning Generalization
Revisiting Token Compression for Accelerating ViT-based Sparse Multi-View 3D Object Detectors
Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learning
Revisiting Visual Corruptions in LVLMs: A Shape–Texture Perspective on Model Failures
REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding
Reviving ConvNeXt for Efficient Convolutional Diffusion Models
RE-VLM: Event-Augmented Vision-Language Model for Scene Understanding
RewardFlow: Generate Images by Optimizing What You Reward
Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation
Reward Sharpness-Aware Fine-Tuning for Diffusion Models
ReWeaver: Towards Simulation-Ready and Topology-Accurate Garment Reconstruction
Rewis3d: Reconstruction for Weakly-Supervised Semantic Segmentation
Rewis3d: Reconstruction Improves Weakly-Supervised Semantic Segmentation
RF4D:Neural Radar Fields for Novel View Synthesis in Outdoor Dynamic Scenes
RFDM: Residual Flow Diffusion Models for Video Editing
RGB-Event based Pedestrian Attribute Recognition: A Benchmark Dataset and An Asymmetric RWKV Fusion Framework

« prev Page 53/74 next »