应用 · computer vision论文趋势 2023–2026
应用方向：computer_vision
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 2,538 篇被归入该主题的论文。
主要结论
- 应用 · computer vision论文从 2023 年的 106 篇增长到 2026 年的 1,326 篇，三年间扩大到约 13 倍。
- ICLR 是发应用 · computer vision论文最多的会议，2023–2026 年共 1,048 篇，占四个会议总量的 41%。
- 交叉方向里涨得最快的是应用 · speech and audio：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 292 篇。
- 在「应用领域」的 10 个主题里，它的 2026 年论文数排第 1；同组增长最快的应用 · health medicine是 61 倍，应用 · computer vision是 13 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 34 | 20 | 8 | 44 | 106 |
| 2024 | 89 | 152 | 25 | 68 | 334 |
| 2025 | 165 | 338 | 201 | 68 | 772 |
| 2026 | 168 | 538 | 554 | 66 | 1,326 |
代表论文
按年份和会议均匀挑选，方便快速看到每一届的样貌
- NeurIPS 2026Adapting Vision Transformers to Organoid Imaging
- ICML 2026$\sigma$: Sigmoid Modulation for Ultra High Resolution Diffusion
- ICLR 2026A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
- CVPR 2026Benchmarking PhD-Level Coding in 3D Geometric Computer Vision
- NeurIPS 2025BiggerGait: Unlocking Gait Recognition with Layer-wise Representations from Large Vision Models
- ICML 2025Action Dubber: Timing Audible Actions via Inflectional Flow
- ICLR 2025A Multiscale Frequency Domain Causal Framework for Enhanced Pathological Analysis
- CVPR 2025BHViT: Binarized Hybrid Vision Transformer
- NeurIPS 2024A Taxonomy of Challenges to Curating Fair Datasets
- ICML 2024Enhancing Vision Transformer: Amplifying Non-Linearity in Feedforward Network Module
- ICLR 2024Boosting Vanilla Lightweight Vision Transformers via Re-parameterization
- CVPR 2024Degrees of Freedom Matter: Inferring Dynamics from Point Trajectories
- NeurIPS 2023[Re] Masked Autoencoders Are Small Scale Vision Learners: A Reproduction Under Resource Constraints
- ICML 2023Can Forward Gradient Match Backpropagation?
- ICLR 2023Fundamental limits on the robustness of image classifiers
- CVPR 2023BiFormer: Vision Transformer With Bi-Level Routing Attention
- NeurIPS 2026Computer Use at the Edge of the Statistical Precipice
- ICML 20263D Scene Assertion Verification
- ICLR 2026A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
- CVPR 2026Describe Anything Anywhere At Any Moment
全部应用 · computer vision论文
共 772 篇，第 11 / 16 页
- ICML 2025History-Guided Video Diffusion
- ICML 2025How Do Images Align and Complement LiDAR? Towards a Harmonized Multi-modal 3D Panoptic Segmentation
- ICML 2025How Effective Can Dropout Be in Multiple Instance Learning ?
- ICML 2025Human Body Restoration with One-Step Diffusion Model and A New Benchmark
- ICML 2025Hybrid Spiking Vision Transformer for Object Detection with Event Cameras
- ICML 2025HybridGS: High-Efficiency Gaussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder
- ICML 2025Impossible Videos
- ICML 2025Improving Compositional Generation with Diffusion Models Using Lift Scores
- ICML 2025Improving LLM Video Understanding with 16 Frames Per Second
- ICML 2025InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective
- ICML 2025Instruct2See: Learning to Remove Any Obstructions Across Distributions
- ICML 2025IntLoRA: Integral Low-rank Adaptation of Quantized Diffusion Models
- ICML 2025IRBridge: Solving Image Restoration Bridge with Pre-trained Generative Diffusion Models
- ICML 2025L-Diffusion: Laplace Diffusion for Efficient Pathology Image Segmentation
- ICML 2025Large Displacement Motion Transfer with Unsupervised Anytime Interpolation
- ICML 2025Learning Adaptive Lighting via Channel-Aware Guidance
- ICML 2025Learning Attribute-Aware Hash Codes for Fine-Grained Image Retrieval via Query Optimization
- ICML 2025Learning Event Completeness for Weakly Supervised Video Anomaly Detection
- ICML 2025Learning Invariant Causal Mechanism from Vision-Language Models
- ICML 2025LieRE: Lie Rotational Positional Encodings
- ICML 2025LightningDrag: Lightning Fast and Accurate Drag-based Image Editing Emerging from Videos
- ICML 2025Lightweight-Mark: Rethinking Deep Learning-Based Watermarking
- ICML 2025LLaVA-ReID: Selective Multi-image Questioner for Interactive Person Re-Identification
- ICML 2025LOCATE 3D: Real-World Object Localization via Self-Supervised Learning in 3D
- ICML 2025Long-Term TalkingFace Generation via Motion-Prior Conditional Diffusion Model
- ICML 2025LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression
- ICML 2025LRA-QViT: Integrating Low-Rank Approximation and Quantization for Robust and Efficient Vision Transformers
- ICML 2025MaskTwins: Dual-form Complementary Masking for Domain-Adaptive Image Segmentation
- ICML 2025MiraGe: Editable 2D Images using Gaussian Splatting
- ICML 2025MIRROR: Make Your Object-Level Multi-View Generation More Consistent with Training-Free Rectification
- ICML 2025MoMa: Modulating Mamba for Adapting Image Foundation Models to Video Recognition
- ICML 2025More Than Meets the Eye: Enhancing Multi-Object Tracking Even with Prolonged Occlusions
- ICML 2025MuLan: Adapting Multilingual Diffusion Models for Hundreds of Languages with Negligible Cost
- ICML 2025Multi-Modal Object Re-identification via Sparse Mixture-of-Experts
- ICML 2025Mutual Learning for SAM Adaptation: A Dual Collaborative Network Framework for Source-Free Domain Transfer
- ICML 2025Not All Tokens Matter All The Time: Dynamic Token Aggregation Towards Efficient Detection Transformers
- ICML 2025Objective drives the consistency of representational similarity across datasets
- ICML 2025One Diffusion Step to Real-World Super-Resolution via Flow Trajectory Distillation
- ICML 2025One-dimensional Path Convolution
- ICML 2025Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models
- ICML 2025Origin Identification for Text-Guided Image-to-Image Diffusion Models
- ICML 2025Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection
- ICML 2025Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splatting
- ICML 2025PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis
- ICML 2025PhySpec: Physically Consistent Spectral Reconstruction via Orthogonal Subspace Decomposition and Self-Supervised Meta-Auxiliary Learning
- ICML 2025PiD: Generalized AI-Generated Images Detection with Pixelwise Decomposition Residuals
- ICML 2025PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop
- ICML 2025Playmate: Flexible Control of Portrait Animation via 3D-Implicit Space Guided Diffusion
- ICML 2025Point Cloud Dataset Distillation
- ICML 2025Pre-training Auto-regressive Robotic Models with 4D Representations
在 Claude、Cursor 里直接问这个趋势
接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。
可以直接复制这句话问 AI
用 paperlist 查应用 · computer vision在 ICLR 2023–2026 的论文趋势，哪一年涨得最快？
AI 会调用的工具：get_topic_stats {"topic": "app-computer-vision"}
常见问题
- 2023–2026 年四大会议一共有多少篇应用 · computer vision论文？
- 共 2,538 篇，按年份分别是 2023 年 106 篇、2024 年 334 篇、2025 年 772 篇、2026 年 1,326 篇。
- NeurIPS 2026 有多少篇应用 · computer vision论文？
- NeurIPS 2026 有 168 篇应用 · computer vision论文，上一届是 165 篇。
- ICLR 2026 有多少篇应用 · computer vision论文？
- ICLR 2026 有 538 篇应用 · computer vision论文，上一届是 338 篇。
- ICML 2026 有多少篇应用 · computer vision论文？
- ICML 2026 有 554 篇应用 · computer vision论文，上一届是 201 篇。
- CVPR 2026 有多少篇应用 · computer vision论文？
- CVPR 2026 有 66 篇应用 · computer vision论文，上一届是 68 篇。
- 哪个会议的应用 · computer vision论文最多？
- ICLR，2023–2026 年共 1,048 篇。