CVPR 2023

3,088 papers

Top Research Directions

Research directions most represented by CVPR 2023 papers.

Generalizing from Synthetic Data to the Real World 39
Aligning Global Semantics with Local Spatial Precision 34
Mitigating Error Propagation in Pseudo Labeling 32
Accelerating Neural Scene Representations for Real Time Rendering 23
Mitigating Multi View Inconsistency in Generative 3D Modeling 22
Eliminating Per-Scene Optimization with Feed-Forward 3D Reconstruction 22
Overcoming the Static Scene Assumption 21
Exploiting Temporal Redundancy in Video Neural Networks 21
Material Lighting Disentanglement for 3D Reconstruction 21
Bypassing the Computational Bottleneck of Adversarial Training 21
Resolving Shape Radiance Ambiguity in Sparse View Reconstruction 20
Overcoming the Iterative Inference Latency of Diffusion Models 19
Tri-Plane Feature Representations for 3D Deep Learning 18
Deep Learning on Unstructured 3D Point Clouds 17
Resolving Depth Ambiguity in Monocular 3D Reconstruction 17

Top Authors

AuthorPapers
Yu Qiao23
Ying Shan19
Luc Van Gool15
Yi-Zhe Song14
Yongdong Zhang14
Dacheng Tao13
Zhiwei Xiong13
Hongsheng Li12
Jiajun Wu12
Leonidas Guibas12
Michael J. Black12
Yi Yang12
Jie Zhou11
Otmar Hilliges11
Wanli Ouyang11
Yu-Gang Jiang11
Deva Ramanan10
Fahad Shahbaz Khan10
Han Hu10
Jiaya Jia10

Impactful Papers

Papers from CVPR 2023 that shaped research clusters.

Foundational

Influential

Papers

Title
TimeBalance: Temporally-Invariant and Temporally-Distinctive Video Representations for Semi-Supervised Action Recognition
TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding
TINC: Tree-Structured Implicit Neural Compression
TinyMIM: An Empirical Study of Distilling MIM Pre-trained Models
TIPI: Test Time Adaptation with Transformation Invariance
TMO: Textured Mesh Acquisition of Objects with a Mobile Device by using Differentiable Rendering
Token Boosting for Robust Self-Supervised Visual Transformer Pre-training
Token Contrast for Weakly-Supervised Semantic Segmentation
TokenHPE: Learning Orientation Tokens for Efficient Head Pose Estimation via Transformers
Token Turing Machines
TopDiG: Class-agnostic Topological Directional Graph Extraction from Remote Sensing Images
Top-Down Visual Attention from Analysis by Synthesis
TOPLight: Lightweight Neural Networks with Task-Oriented Pretraining for Visible-Infrared Recognition
TopNet: Transformer-Based Object Placement Network for Image Compositing
Topology-Guided Multi-Class Cell Context Generation for Digital Pathology
Total Selfie: Generating Full-Body Selfies
ToThePoint: Efficient Contrastive Learning of 3D Point Clouds via Recycling
Toward Accurate Post-Training Quantization for Image Super Resolution
Toward RAW Object Detection: A New Benchmark and A New Model
Towards accurate image coding: Improved autoregressive image generation with dynamic vector quantization
Towards All-in-One Pre-Training via Maximizing Multi-Modal Mutual Information
Towards Artistic Image Aesthetics Assessment: a Large-scale Dataset and a New Method
Towards a Smaller Student: Capacity Dynamic Distillation for Efficient Image Retrieval
Towards Automatic Power Battery Detection: New Challenge, Benchmark Dataset and Baseline
Towards Benchmarking and Assessing Visual Naturalness of Physical World Adversarial Attacks
Towards Better Decision Forests: Forest Alternating Optimization
Towards Better Gradient Consistency for Neural Signed Distance Functions via Level Set Alignment
Towards Better Stability and Adaptability: Improve Online Self-Training for Model Adaptation in Semantic Segmentation
Towards Bridging the Performance Gaps of Joint Energy-Based Models
Towards Building Self-Aware Object Detectors via Reliable Uncertainty Quantification and Calibration
Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations
Towards Domain Generalization for Multi-view 3D Object Detection in Bird-Eye-View
Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognition
Towards Effective Visual Representations for Partial-Label Learning
Towards Efficient Use of Multi-Scale Features in Transformer-Based Object Detectors
Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers
Towards Fast Adaptation of Pretrained Contrastive Models for Multi-channel Video-Language Retrieval
Towards Flexible Multi-modal Document Models
Towards Generalisable Video Moment Retrieval: Visual-Dynamic Injection to Image-Text Pre-Training
Towards High-Quality and Efficient Video Super-Resolution via Spatial-Temporal Data Overfitting
Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training
Towards Learning a Generalist Model for Embodied Navigation
Towards Modality-Agnostic Person Re-identification with Descriptive Query
Towards More Unified In-Context Visual Understanding
Towards Open-World Segmentation of Parts
Towards Practical Plug-and-Play Diffusion Models
Towards Professional Level Crowd Annotation of Expert Domain Data
Towards Realistic Long-Tailed Semi-Supervised Learning: Consistency is All You Need
Towards Robust Tampered Text Detection in Document Image: New Dataset and New Solution
Towards Scalable 3D Anomaly Detection and Localization: A Benchmark via 3D Anomaly Synthesis and A Self-Supervised Learning Network

« prev Page 56/62 next »