CVPR 2020 已經(jīng)公布了大多數(shù)workshop的細節(jié),與這些workshop對應(yīng)的還有很多計算機視覺算法比賽,這些比賽代表著或新興、或?qū)嵱?、或有趣、或被忽略的研究方向?br>一文匯總了大部分比賽(因有些還未公布網(wǎng)址),大多數(shù)已經(jīng)開賽放出了數(shù)據(jù)集,歡迎大家按圖索驥。WebVision Image Classification技術(shù)方向:圖像分類、半監(jiān)督學(xué)習(xí)、無監(jiān)督學(xué)習(xí)http://www.vision.ee./webvision/challenge.htmlChallenges for Computer Vision in Agriculturehttps://www./prize-challengeThermal Image Super-Resolution Challengehttp:///pbvs/20/challenge.htmlLow-rate and P-frame compression challengeshttp://challenge./motivation/Diagram Image Retrieval and Analysis (DIRA) Challengehttp://cvpr-dira./challenge/Low-Power Computer Vision Competition技術(shù)方向:低功耗計算機視覺、目標(biāo)檢測、圖像分類http://cbcsl.ece./enc-2020/index.htmlImage Matching: Local Features and Beyond技術(shù)方向:局部特征提取、含噪數(shù)據(jù)擬合https://image-matching-workshop./ps.深度學(xué)習(xí)時代,難得還有這么傳統(tǒng)的方向。New Trends in Image Restoration and Enhancement Challenges (NTIRE)技術(shù)方向:圖像超分辨率、圖像去噪、去模糊、去摩爾紋、重建、去霧http://www.vision.ee./ntire20/技術(shù)方向:低質(zhì)圖像增強、圖像恢復(fù)、目標(biāo)檢測、人臉驗證ISIC Skin Image Analysis Workshop技術(shù)方向:皮膚圖像分析、圖像分割、圖像分類Weakly Supervised Learning Challenges技術(shù)方向:弱監(jiān)督學(xué)習(xí)、圖像分割、場景解析、目標(biāo)定位https://lidchallenge./challenge.htmlhttps://www./earthvision2020/challenge.htmlhttps:///workshops/bmtt2020/tracking.htmlDAVIS Video Object Segmentationhttps:///challenge2020/index.html技術(shù)方向:智能交通、車輛檢測、卡口車輛計數(shù)、重識別、車輛跟蹤、交通異常檢測https://www./2020-challenge-tracks/https:///tasks-and-datasets/image-captioning/VQA and Dialog ChallengesContinual Learning Challenge技術(shù)方向:持續(xù)學(xué)習(xí)https://sites.google.com/view/clvision2020/challengeCross-Domain Few-Shot Learning (CD-FSL) Challenge技術(shù)方向:跨域少樣本學(xué)習(xí)DeepFashion2 and FashionIQ Challengeshttps://sites.google.com/view/cvcreative2020#h.p_Qypem5p_Nm6cThe Seventh Workshop on Fine-Grained Visual Categorizationhttps://sites.google.com/view/fgvc7Challenge on Remote Physiological Signal Sensinghttps://competitions./competitions/22287Localization, Odometry, SLAM Challengehttps://sites.google.com/view/vislocslamcvpr2020/home2nd ScanNet Indoor Scene Understanding Challengehttp://www./cvpr2020workshop/ 未完待續(xù),歡迎持續(xù)關(guān)注公眾號后續(xù)文章。
|