CVPR 2022 57 篇论文分方向整理 + 打包下载|涵盖目标检测、语义分割、人群计数、异常检测等方向
CVPR2022論文速遞系列:
CVPR 2022 3月3日論文速遞(22 篇打包下載)涵蓋網絡架構設計、姿態估計、三維視覺、動作檢測、語義分割等方向
CVPR 2022 3月4日論文速遞(29 篇打包下載)涵蓋目標檢測、全景分割、異常檢測、度量學習、對比學習、目標跟蹤等方向
CVPR 2022 3月7日論文速遞(17 篇打包下載)涵蓋 3D 目標檢測、醫學影像、圖像去模糊、車道線檢測等方向
推薦閱讀:CVPR 2022 最全整理:論文分方向匯總 / 代碼 / 解讀 / 直播 / 項目(更新中)【計算機視覺】
以下為3月9日更新的57篇CVPR2022論文分方向整理,打包下載: 下載地址
2D 目標檢測(2D Object Detection)
【1】Unknown-Aware Object Detection: Learning What You Don’t Know from Videos in the Wild(未知感知對象檢測:從野外視頻中學習你不知道的東西)
paper | code
偽裝目標檢測(Camouflaged Object Detection)
【1】Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection(放大和縮小:用于偽裝目標檢測的混合尺度三元組網絡)
paper | code
超分辨率(Super Resolution)
【1】Reflash Dropout in Image Super-Resolution(圖像超分辨率中的閃退dropout)
paper
【2】Towards Bidirectional Arbitrary Image Rescaling: Joint Optimization and Cycle Idempotence(邁向雙向任意圖像縮放:聯合優化和循環冪等)
paper
【3】【HyperTransformer: A Textural and Spectral Feature Fusion Transformer for Pansharpening(用于全色銳化的紋理和光譜特征融合Transformer)
paper | code
小樣本學習/零樣本學習(Few-shot Learning/Zero-shot Learning)
【1】Learning to Affiliate: Mutual Centralized Learning for Few-shot Classification(小樣本分類的相互集中學習)
paper
【2】MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning(用于零樣本學習的相互語義蒸餾網絡)
keywords: Zero-Shot Learning, Knowledge Distillation
paper | code
神經網絡可解釋性(Neural Network Interpretability)
【1】Interpretable part-whole hierarchies and conceptual-semantic relationships in neural networks(神經網絡中可解釋的部分-整體層次結構和概念語義關系)
paper
姿態估計(Human Pose Estimation)
【1】Forecasting Characteristic 3D Poses of Human Actions()
paper | project | video
醫學影像(Medical Imaging)
【1】Adaptive Early-Learning Correction for Segmentation from Noisy Annotations(從噪聲標簽中分割的自適應早期學習校正)
keywords: medical-imaging segmentation, Noisy Annotations
paper | code
光流/位姿/運動估計(Optical Flow/Pose/Motion Estimation)
【1】CPPF: Towards Robust Category-Level 9D Pose Estimation in the Wild(CPPF:在野外實現穩健的類別級 9D 位姿估計)
paper | code
【2】OVE6D: Object Viewpoint Encoding for Depth-based 6D Object Pose Estimation(用于基于深度的 6D 對象姿態估計的對象視點編碼)
paper | code
【3】CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimation(用于聯合光流和場景流估計的雙向相機-LiDAR 融合)
paper
視覺語言表征學習(Vision-language Representation Learning)
【1】L-Verse: Bidirectional Generation Between Image and Text(圖像和文本之間的雙向生成) (Oral Presentation)
paper
點云(Point Cloud)
【1】Shape-invariant 3D Adversarial Point Clouds(形狀不變的 3D 對抗點云)
paper | code
【2】ART-Point: Improving Rotation Robustness of Point Cloud Classifiers via Adversarial Rotation(通過對抗旋轉提高點云分類器的旋轉魯棒性)
paper
【3】Lepard: Learning partial point cloud matching in rigid and deformable scenes(Lepard:在剛性和可變形場景中學習部分點云匹配)
paper | code
深度估計(Depth Estimation)
【1】Rethinking Depth Estimation for Multi-View Stereo: A Unified Representation and Focal Loss(重新思考多視圖立體的深度估計:統一表示和焦點損失)
paper | code
場景重建/新視角合成(Novel View Synthesis)
【1】Point-NeRF: Point-based Neural Radiance Fields(基于點的神經輻射場)
paper | code |project
圖像生成/圖像合成/視頻合成(Image Generation/Image Synthesis/Video Generation)
【1】Exploring Dual-task Correlation for Pose Guided Person Image Generation(探索姿勢引導人物圖像生成的雙任務相關性)
paper | code
【2】Show Me What and Tell Me How: Video Synthesis via Multimodal Conditioning(告訴我什么并告訴我如何:通過多模式調節進行視頻合成)
paper | code
風格遷移(Style Transfer)
【1】How Well Do Sparse Imagenet Models Transfer?(稀疏 Imagenet 模型的遷移效果如何?)
paper
【2】Style-ERD: Responsive and Coherent Online Motion Style Transfer(響應式和連貫的在線運動風格遷移)
paper
人群計數(Crowd Counting)
【1】Boosting Crowd Counting via Multifaceted Attention(通過多方面注意提高人群計數)
paper | code
數據集(Dataset)
【1】Kubric: A scalable dataset generator(Kubric:可擴展的數據集生成器)
paper | code
【2】A Large-scale Comprehensive Dataset and Copy-overlap Aware Evaluation Protocol for Segment-level Video Copy Detection(用于分段級視頻復制檢測的大規模綜合數據集和復制重疊感知評估協議)
VCSL (Video Copy Segment Localization) dataset
paper | dataset, metric and benchmark codes
語義分割(Semantic Segmentation)
【1】Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels(使用不可靠偽標簽的半監督語義分割)
paper | code | project
【2】Weakly Supervised Semantic Segmentation using Out-of-Distribution Data(使用分布外數據的弱監督語義分割)
paper | code
【3】Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentation(弱監督語義分割的自監督圖像特定原型探索)
paper | code
【4】Multi-class Token Transformer for Weakly Supervised Semantic Segmentation(用于弱監督語義分割的多類token Transformer)
paper | code
【5】Cross Language Image Matching for Weakly Supervised Semantic Segmentation(用于弱監督語義分割的跨語言圖像匹配)
paper
【6】Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers(從注意力中學習親和力:使用 Transformers 的端到端弱監督語義分割)
paper | code
實例分割(Instance Segmentation)
【1】 E2EC: An End-to-End Contour-based Method for High-Quality High-Speed Instance Segmentation(一種基于端到端輪廓的高質量高速實例分割方法)
paper | code
模型訓練/泛化(Model Training/Generalization)
【1】Towards Efficient and Scalable Sharpness-Aware Minimization(邁向高效和可擴展的銳度感知最小化)
keywords: Sharp Local Minima, Large-Batch Training
paper
行為識別/動作識別/檢測/分割/定位(Action/Activity Recognition)
【1】End-to-End Semi-Supervised Learning for Video Action Detection(視頻動作檢測的端到端半監督學習)
paper
【2】Learnable Irrelevant Modality Dropout for Multimodal Action Recognition on Modality-Specific Annotated Videos(模態特定注釋視頻上多模態動作識別的可學習不相關模態丟失)
paper
【3】Weakly Supervised Temporal Action Localization via Representative Snippet Knowledge Propagation(通過代表性片段知識傳播的弱監督時間動作定位)
paper | code
圖像分類(Image Classification)
【1】 GlideNet: Global, Local and Intrinsic based Dense Embedding NETwork for Multi-category Attributes Prediction(用于多類別屬性預測的基于全局、局部和內在的密集嵌入網絡)
keywords: multi-label classification
paper | code | project
聯邦學習(Federated Learning)
【1】 Differentially Private Federated Learning with Local Regularization and Sparsification(局部正則化和稀疏化的差分私有聯邦學習)
paper
GAN/生成式/對抗式(GAN/Generative/Adversarial)
【1】Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon(陰影可能很危險:自然現象的隱秘而有效的物理世界對抗性攻擊)
paper
【2】Protecting Facial Privacy: Generating Adversarial Identity Masks via Style-robust Makeup Transfer(保護面部隱私:通過風格穩健的化妝轉移生成對抗性身份面具)
paper
【3】Adversarial Texture for Fooling Person Detectors in the Physical World(物理世界中愚弄人探測器的對抗性紋理)
paper
三維重建(3D Reconstruction)
【1】Neural Face Identification in a 2D Wireframe Projection of a Manifold Object(流形對象的二維線框投影中的神經人臉識別)
paper | [code](https://manycore- research.github.io/faceformer) | project
【2】Generating 3D Bio-Printable Patches Using Wound Segmentation and Reconstruction to Treat Diabetic Foot Ulcers()
keywords: semantic segmentation, 3D reconstruction, 3D bio-printers
paper
異常檢測(Anomaly Detection)
【1】Generative Cooperative Learning for Unsupervised Video Anomaly Detection(用于無監督視頻異常檢測的生成式協作學習)
paper
持續學習(Continual Learning/Life-long Learning)
【1】On Generalizing Beyond Domains in Cross-Domain Continual Learning(關于跨域持續學習中的域外泛化)
paper
視覺預測(Vision-based Prediction)
【1】Motron: Multimodal Probabilistic Human Motion Forecasting(多模式概率人體運動預測)
paper
對比學習(Contrastive Learning)
【1】Selective-Supervised Contrastive Learning with Noisy Labels(帶有噪聲標簽的選擇性監督對比學習)
paper | code
遷移學習(Transfer Learning)
【1】A Simple Multi-Modality Transfer Learning Baseline for Sign Language Translation(用于手語翻譯的簡單多模態遷移學習基線)
paper
圖像特征提取與匹配(Image feature extraction and matching)
【1】 Probabilistic Warp Consistency for Weakly-Supervised Semantic Correspondences(弱監督語義對應的概率扭曲一致性)
paper | code
量化(Quantization)
【1】IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization(學習具有類內異質性的合成圖像以進行零樣本網絡量化)
paper | code
數據增廣(Data Augmentation)
【1】TeachAugment: Data Augmentation Optimization Using Teacher Knowledge(使用教師知識進行數據增強優化)
paper | code
Transformer
【1】Delving Deep into the Generalization of Vision Transformers under Distribution Shifts(深入研究分布變化下的視覺Transformer的泛化)
keywords: out-of-distribution (OOD) generalization, Vision Transformers
paper | code
CNN
【1】DeltaCNN: End-to-End CNN Inference of Sparse Frame Differences in Videos(視頻中稀疏幀差異的端到端 CNN 推斷)
keywords: sparse convolutional neural network, video inference accelerating
paper
暫無分類
[5] Contrastive Conditional Neural Processes(對比條件神經過程)
paper
[4] Deep Rectangling for Image Stitching: A Learning Baseline(圖像拼接的深度矩形:學習基線)(Image Stitching)
paper | code
[3] Online Learning of Reusable Abstract Models for Object Goal Navigation(對象目標導航可重用抽象模型的在線學習)
paper
總結
以上是生活随笔為你收集整理的CVPR 2022 57 篇论文分方向整理 + 打包下载|涵盖目标检测、语义分割、人群计数、异常检测等方向的全部內容,希望文章能夠幫你解決所遇到的問題。
- 上一篇: CVPR 2022 3月7日论文速递(1
- 下一篇: CVPR 2022|从原理和代码详解FA