关注公众号,发现CV技术之美
CVPR 每年都会举办一系列竞赛,本文汇总了 CVPR 2024 举办的所有竞赛,共计43个(部分已截止)。
这些竞赛涵盖了计算机视觉领域的多个方面,包括三维重建、图像字幕、人工智能、图像生成、视频理解等。其中,一些竞赛延续了以往几年的主题,鼓励研究人员刷新旧榜;另一些竞赛则以全新的主题,以此拓展计算机视觉研究的领域。
1.OpenSUN 3D:2nd Workshop on Open-Vocabulary 3D Scene Understanding
项目主页:https://opensun3d.github.io/
技术方向:三维场景理解
赛道一:Open-vocabulary 3D Object Instance Search
赛道二:Open-vocabulary 3D Functionality Grounding
截止日期:2024 年 6 月 9 日
2.2nd Workshop on Compositional 3D Vision
项目主页:https://3dcompat-dataset.org/workshop/C3DV24/
技术方向:粗粒度和细粒度三维分割
已截止
3.(3rd) Monocular Depth Estimation Challenge
项目主页:https://jspenmar.github.io/MDEC/
技术方向:单目深度估计
已截止

4.10th International Workshop on Computer Vision in Sports
项目主页:https://vap.aau.dk/cvsports/
技术方向:体育视频分析
赛道一:Ball Action Spotting
赛道二:Dense Video Captioning
赛道三:Multi-View Foul Recognition
赛道四:Game State Reconstruction
最佳论文可获得 1000 美元奖金
已截止

5.Agriculture-Vision: Challenges & Opportunities for Computer Vision in Agriculture
项目主页:https://www.agriculture-vision.com/agriculture-vision-2024/prize-challenge-2024
技术方向:航空图像农业模式识别算法
本次挑战赛核心是使用扩展的、未标记的农业视觉数据集和已标记的原始农业视觉数据集。通过有效应用半监督学习技术来提高模型性能。
总奖金 5000 美元
已截止
6.AIS: Vision, Graphics and AI for Streaming
项目主页:https://ai4streaming-workshop.github.io/
技术方向:图像/视频应用研究
赛道一:Real-time Compressed Image Super-Resolution (已截止)
赛道二:UGC Video Quality Assessment (已截止)
赛道三:Event-based Eye Tracking (已截止)
赛道四:Mobile Real-time Video Super-Resolution (已截止)
赛道五:Efficient Video Super-Resolution (已截止)
赛道六:Depth Upsampling and Refinement (2024 年 7 月 24 日截止)
奖金 10,000 美元+PS5

7.MetaFood Workshop (MTF)
项目主页:https://sites.google.com/view/cvpr-metafood-2024
技术方向:三维食物重建
本次挑战赛旨在将三维重建技术应用于动态和不受控的环境,例如在进食过程中捕捉食物图像,从而推动三维重建技术的发展。
第一阶段提交截止日期:2024 年 5 月 24 日
第二阶段提交截止日期:2024 年 5 月 27 日
总奖金 1000 美元
8.RetailVision - Field Overview and Amazon Deep Dive
项目主页:https://retailvisionworkshop.github.io/
技术方向:实体店购物场景研究
本次挑战赛有两个赛道,一个是动作识别赛道,另一个是产品验证赛道。在动作识别场景中,模型检测购物者的动作,如将产品放入购物篮;验证轨道则是识别产品的身份。
赛道一:Video Temporal Action Localization (TAL) and Spatial Temporal Action Localization (STAL)
赛道二:Multi-modal Product Retrieval (MPR)

截止日期:5 月 31 日
9.AVA: Accessibility, Vision and Autonomy Meet
项目主页:https://accessibility-cv.github.io/
技术方向:自主系统与残疾行人交互研究
赛道一:CVPR2024 AVA Accessibility Vision and Autonomy Challenge - Keypoint Track
赛道二:CVPR2024 AVA Accessibility Vision and Autonomy Challenge - Segmentation Track
截止日期:6月11日
最佳团队可获得 500 美元资金奖励和一台 OAK-D 摄像机

10.VizWiz Grand Challenge: Describing Images and Videos Taken by Blind People
项目主页:https://vizwiz.org/workshops/2024-vizwiz-grand-challenge-workshop/
技术方向:视觉问答、图像分类、物体识别
赛道一Visual Question Answering
:Predict Answer to a Visual Question
:Predict Answerability of a Visual Question
Answer Grounding
single answer grounding recognition
few-shot video object recognition
Few-Shot Private Object Localization
zero-shot image classification

已截止
11.L3D-IVU: 3rd Workshop on Learning with Limited Labelled Data for Image and Video Understanding
项目主页:https://sites.google.com/view/l3divu2024/overview
技术方向:遥感图像理解
第一名:1000 美元
已截止

12.New frontiers for zero-shot Image Captioning Evaluation (NICE)
项目主页:https://nice.lgresearch.ai/
技术方向:零样本图像字幕
赛道一:Zero-shot image captioning evaluation
赛道二:Caption re-ranking evaluation
每个赛道冠军 5,000 美元奖金,亚军 2,500 美元
已截止

13.7th Workshop on Autonomous Driving (WAD)
项目主页:https://cvpr2024.wad.vision/
技术方向:自动驾驶
赛道一:Announcing the 2024 Waymo Open Dataset Challenges
赛道二:Argoverse 2024 Challenges
已截止

14.2nd Face Recognition Challenge in the Era of Synthetic Data (FRCSyn)
项目主页:https://frcsyn.github.io/CVPR2024.html
技术方向:人脸识别
任务一:synthetic data for demographic bias mitigation
任务二:synthetic data for overall performance improvement
已截止

15.2nd Workshop and Challenge on DeepFake Analysis and Detection
项目主页:https://www.dfad.unimore.it/
技术方向:DeepFake分析检测截止日期:6 月 1 日

16.6th Workshop and Competition on Affective Behavior Analysis in-the-wild
项目主页:https://affective-behavior-analysis-in-the-wild.github.io/6th/
技术方向:情感行为分析
赛道一:Valence-Arousal Estimation Challenge
赛道二:Expression Recognition Challenge
赛道三:Action Unit Detection Challenge
赛道四:Compound Expression Recognition Challenge
赛道五:Emotional Mimicry Intensity Estimation Challenge
已截止


17.The 5th Face Anti-Spoofing Workshop
项目主页:https://sites.google.com/view/face-anti-spoofing-challenge/welcome/challengecvpr2024
技术方向:人脸活体检测
赛道一:Unified Physical-Digital Face Attack Detection
赛道二:Snapshot Spectral Imaging Face Anti-spoofing
已截止

18.20th Workshop on Perception Beyond the Visible Spectrum
项目主页:https://pbvs-workshop.github.io/index.html
技术方向:图像超分辨率
赛道一:5th Thermal Image Super-Resolution Challenge (TISR)
PBVS 2024 Thermal Image Super-Resolution Challenge (TISR)
PBVS 2024 Thermal Image Super-Resolution Challenge (GTISR)
赛道二:Multi-modal Aerial View Imagery Challenge: Classification (MAVIC-C)
赛道三:Multi-modal Aerial View Imagery Challenge: Translation (MAVIC-T)
已截止

19.9th New Trends in Image Restoration and Enhancement Workshop and Challenges
项目主页:https://cvlai.net/ntire/2024/
技术方向:图像/视频处理、修复和增强
NTIRE 2024 challenges
Dense and Non-Homogeneous Dehazing
Night Photography Rendering
Blind Compressed Image Enhancement
Shadow Removal - Track 1 Fidelity
Shadow Removal - Track 2 Perceptual
Efficient Super Resolution
Image Super Resolution (x4)
Light Field Image Super-Resolution - Track 1 Fidelity
Light Field Image Super-Resolution - Track 2 Efficiency
Stereo Image Super-Resolution - Track 1 Bicubic
Stereo Image Super-Resolution - Track 2 Realistic
HR Depth from Images of Specular and Transparent Surfaces - Track 1 Stereo
HR Depth from Images of Specular and Transparent Surfaces - Track 2 Mono
Bracketing Image Restoration and Enhancement - Track 1
Bracketing Image Restoration and Enhancement - Track 2
Portrait Quality Assessment
Quality Assessment for AI-Generated Content - Track 1 Image
Quality Assessment for AI-Generated Content - Track 2 Video
Restore Any Image Model (RAIM) in the Wild
RAW Image Super-Resolution
Short-form UGC Video Quality Assessment
Low Light Enhancement
RAW Burst Alignment and ISP Challenge
AIS: Vision, Graphics and AI for Streaming 2024 challenges
Real-time Compressed Image Super-Resolution
Real-time Video Super-Resolution
UGC Video Quality Assessment
Event-based Eye Tracking
已截止
20.The 5th Omnidirectional Computer Vision Workshop
项目主页:https://sites.google.com/view/omnicv2024/home
技术方向:全景成像
TBD

21.The 7th Workshop and Challenge Bridging the Gap between Computational Photography and Visual Recognition (UG2+)
项目主页:https://cvpr2024ug2challenge.github.io/
技术方向:图像重建、恶劣天气下图像语义分割、无人机跟踪
赛道一:ATMOSPHERIC TURBULENCE MITIGATION
赛道二:WEATHERPROOF DATASET CHALLENGE: SEMANTIC SEGMENTATION ON PAIRED REAL DATA
赛道三:UAV TRACKING AND POSE ESTIMATION
已截止
22.First Joint Egocentric Vision (EgoVis) Workshop
项目主页:https://egovis.github.io/cvpr24/
技术方向:可穿戴相机、智能眼镜和 AR/VR 头戴式设备应用研究
挑战一:HoloAssist Challenges(6月6日截止)
挑战二:Aria Digital Twin Challenges(已截止)
挑战三:Aria Synthetic Environments Challenges(已截止)
挑战四:Ego4D Challenges(5月31日截止;其中Ego4D AV Localization赛道5月20日截止)
挑战五:Ego-Exo4D Challenges(5月31日截止)
挑战六:EPIC-Kitchens Challenges(6月1日截止)

23.The 5th Annual Embodied AI Workshop
项目主页:https://embodied-ai.org/#challenges
技术方向:嵌入式人工智能
挑战一:ARNOLD Challenge(TBD)
挑战二:HAZARD Challenge(TBD)
挑战三:HomeRobot OVMM(TBD)
挑战四:ManiSkill-ViTac(已截止)
挑战五:MultiON(6月3日截止)
挑战六:PRS(TBD)
24.5th Workshop on Continual Learning in Computer Vision (CLVISION)
项目主页:https://sites.google.com/view/clvision2024/
技术方向:持续学习
挑战赛:5th CLVISION Challenge (2024)(5月21日截止)
25.2nd Workshop on What is Next in Multimodal Foundation Models?
项目主页:https://sites.google.com/view/2nd-mmfm-workshop/
技术方向:多模态基础模型
挑战赛:MMFM Challenge
第 1 阶段提交截止日期:2024 年 5 月 15 日
第 2 阶段提交截止日期:2024 年 5 月 30 日

26.Foundation Models for Autonomous Systems
项目主页:https://opendrivelab.com/cvpr2024/workshop/
技术方向:自动驾驶
赛道一:end-to-end driving at scale
赛道二:predictive world model
创新奖 18,000 美元
杰出冠军 9,000 美元
荣誉亚军 3,000 美元
6月1日截止

27.The 4th Workshop of Adversarial Machine Learning on Computer Vision: Robustness of Foundation Models
项目主页:https://cvpr24-advml.github.io/
技术方向:黑盒多模态基础模型、对抗性攻击
挑战赛详情 TBD
28.GenAI Media Generation Challenge for Computer Vision Workshop
项目主页:https://gamgc.github.io/
技术方向:图像生成、编辑
赛道一:Text-to-image Generation
赛道二:Text-guided Image Editing
赛道三:Movie Generation Challenge
已截止

29.Rhobin 2024: The second Rhobin challenge on Reconstruction of Human-Object Interaction
项目主页:https://rhobin-challenge.github.io/index.html
技术方向:人机交互
赛道一:3D human reconstruction
赛道二:6DoF pose estimation of rigid objects
赛道三:Joint reconstruction of human and object
赛道四:Video-based tracking
赛道五:3D contact prediction from 2D images
5月30日截止

30.Third Workshop of Mobile Intelligent Photography & Imaging
项目主页:https://mipi-challenge.org/MIPI2024/
技术方向:图像传感器和成像算法
赛道一:Few-shot Raw Image Denoising
赛道二:Demosaic for HybridEVS Camera
赛道三:Nighttime Flare Removal
已截止
31.Multimodal Algorithmic Reasoning Workshop
项目主页:https://marworkshop.github.io/cvpr24/index.html
技术方向:神经算法学习、多模态推理和智能认知模型研究
挑战赛:SMART-101 CVPR 2024 Challenge
6月9日截止
32.VAND 2.0: Visual Anomaly and Novelty Detection
项目主页:https://sites.google.com/view/vand-2-0-cvpr-2024/home
技术方向:视觉异常检测
赛道一:Adapt & Detect: Robust Anomaly Detection in Real-World Applications
赛道二:VLM Anomaly Challenge: Few-Shot Learning for Logical and Structural Detection
截止日期:6 月 1 日

33.Visual Perception via Learning in an Open World
项目主页:https://vplow.github.io/vplow_4th.html
技术方向:视觉感知
挑战一:Object Instance Detection Challenge(6月10截止)
挑战二:Challenge of Exploiting Foundation Models to Annotate Data from Annotation Instructions(6月8日截止)
挑战三:Challenge of Open-Vocabulary Part Segmentation(6月9日截止)
挑战四:Challenge of Vast Vocabulary Visual Detection (V3Det Challenge)(6月3日截止)
34.4th Workshop on Physics Based Vision meets Deep Learning (PBDL2024)
项目主页:https://pbdl-ws.github.io/pbdl2024/index.html
技术方向:低光图像处理、HDR成像
Low-Light Enhancement and Detection Challenge
Low-light Object Detection and Instance Segmentation
Low-light raw video denoising with realistic motion
Low-light SRGB image enhancement
Extreme Low-Light Image Denoising
Low-light Raw Image Enhancement
High Dynamic Range Imaging Challenge
HDR Reconstruction from a Single Raw Image
Highspeed HDR Video Reconstruction from Events
Raw Image Based Over-Exposure Correction
已截止
35.The Sixth Workshop on Deep Learning for Geometric Computing (DLGC 2024)
项目主页:https://sites.google.com/view/dlgc-workshop-cvpr2024/home
技术方向:几何形状分析
挑战一:Breaking Best: A geometric fracture reassembly challenge(6月3日截止)
挑战二:SkelNetOn Challenges(已截止)
Pixel SkelNetOn
Point SkelNetOn
Parametric SkelNetOn
Image SkelNetOn
Breaking Best

36.AI4Space 2024
项目主页:https://aiforspace.github.io/2024/
技术方向:航天器轨迹估计、语义分割
赛道一:Spacecraft Semantic Segmentation
赛道二:Spacecraft Trajectory Estimation
已截止

37.3rd Workshop on Vision Datasets Understanding and DataCV Challenge
项目主页:https://sites.google.com/view/vdu-cvpr24/
技术方向:数据理解
挑战赛:The 2nd CVPR DataCV Challenge
已截止
38.SyntaGen: Harnessing Generative Models for Synthetic Visual Datasets
项目主页:https://syntagen.github.io/
技术方向:生成模型
挑战赛:SyntaGen Competition
截止日期:2024年5月24日

39.1st Workshop on Urban Scene Modeling: Where Vision Meets Photogrammetry and Graphics
项目主页:https://usm3d.github.io/
技术方向:城市场景建模
挑战一:Structured Semantic 3D Reconstruction (S23DR) Challenge(5月28日截止)
挑战二:Building3D Challenge(5月31日截止)

40.4th Workshop and Challenge on Computer Vision in the Built Environment for the Design, Construction, and Operation of Buildings
项目主页:https://cv4aec.github.io/
技术方向:平面图重建和三维建筑模型重建
赛道一:2D Floorplan Reconstruction
赛道二:3D Building Model Reconstruction
6月18日截止
41.8th AI City Challenge
项目主页:https://www.aicitychallenge.org/
技术方向:零售业务和智能交通系统
赛道一:Multi-Camera People Tracking
赛道二:Traffic Safety Description and Analysis
赛道三:Naturalistic Driving Action Recognition
赛道四:Road Object Detection in Fish-Eye Cameras
赛道五:Detecting Violation of Helmet Rule for Motorcyclists
已截止
42.4th International Workshop on Long-form Video Understanding: Towards Multimodal AI Assistant and Copilot
项目主页:https://sites.google.com/view/loveucvpr24
技术方向:长视频理解
竞赛详情 TBD
43.Pixel-level Video Understanding in the Wild Challenge
项目主页:https://www.vspwdataset.com/Workshop2024
技术方向:像素级场景理解
赛道一:Video Semantic Segmentation (VSS)
赛道二:Video Panoptic Segmentation (VPS)
赛道三:Complex Video Object Segmentation
赛道四:Motion Expression guided Video Segmentation
5月15日截止
END
欢迎加入「竞赛/CVPR」交流群👇备注:CVPR