Papers
Object Detection
- R-CNN: Rich feature hierarchies for accurate object detection and semantic segmentation.【链接】
- SPP-Net: Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition.【链接】
- Fast R-CNN: Fast R-CNN.【链接】
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.【链接】
- Mask R-CNN.【链接】
- MultiBox: Scalable Object Detection using Deep Neural Networks.【链接】
- DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection.【链接】
- NoC: Object Detection Networks on Convolutional Feature Maps.【链接】
- DeepBox: Learning Objectness with Convolutional Networks.【链接】
- MR-CNN: Object detection via a multi-region & semantic segmentation-aware CNN model.【链接】
- YOLO: You Only Look Once: Unified, Real-Time Object Detection.【链接】
- YOLOv2: YOLO9000: Better, Faster, Stronger.【链接】
- AttentionNet: Aggregating Weak Directions for Accurate Object Detection.【链接】
- DenseBox: Unifying Landmark Localization with End to End Object Detection.【链接】
- SSD: Single Shot MultiBox Detector.【链接】
- DSSD: Deconvolutional Single Shot Detector.【链接】
- Inside-Outside Net(ION): Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks.【链接】
- G-CNN: An Iterative Grid Based Object Detector.【链接】
- HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection.【链接】
- MultiPathNet: A MultiPath Network for Object Detection.【链接】
- CRAFT: CRAFT Objects from Images.【链接】
- OHEM: Training Region-based Object Detectors with Online Hard Example Mining.【链接】
- R-FCN: Object Detection via Region-based Fully Convolutional Networks.【链接】
- MS-CNN: A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection.【链接】
- PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection.【链接】
- GBD-Net: Gated Bi-directional CNN for Object Detection.【链接】
- StuffNet: Using ‘Stuff’ to Improve Object Detection.【链接】
- Feature Pyramid Network(FPN): Feature Pyramid Networks for Object Detection.【链接】
- CC-Net: Learning Chained Deep Features and Classifiers for Cascade in Object Detection.【链接】
- DSOD: Learning Deeply Supervised Object Detectors from Scratch.【链接】
- NMS: End-to-End Integration of a Convolutional Network, Deformable Parts Model and Non-Maximum Suppression.【链接】
- Weakly Supervised Object Detection: Track and Transfer: Watching Videos to Simulate Strong Human Supervision for Weakly-Supervised Object Detection.【链接】
Object Detection in 3D
- Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient Convolutional Neural Networks.【链接】
Object Detection on RGB-D
Salient Object Detection
- Best Deep Saliency Detection Models (CVPR 2016 & 2015).【链接】
- Large-scale optimization of hierarchical features for saliency prediction in natural images.【链接】
- Predicting Eye Fixations using Convolutional Neural Networks.【链接】
- Saliency Detection by Multi-Context Deep Learning.【链接】
- DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection.【链接】
- SuperCNN: A Superpixelwise Convolutional Neural Network for Salient Object Detection.【链接】
- Shallow and Deep Convolutional Networks for Saliency Prediction.【链接】
- Recurrent Attentional Networks for Saliency Detection.【链接】
- Two-Stream Convolutional Networks for Dynamic Saliency Prediction.【链接】
- Unconstrained Salient Object Detection via Proposal Subset Optimization.【链接】
- DHSNet: Deep Hierarchical Saliency Network for Salient Object Detection.【链接】
- Salient Object Subitizing.【链接】
- Deeply-Supervised Recurrent Convolutional Neural Network for Saliency Detection.【链接】
- Saliency Detection via Combining Region-Level and Pixel-Level Predictions with CNNs.【链接】
- Edge Preserving and Multi-Scale Contextual Neural Network for Salient Object Detection.【链接】
- A Deep Multi-Level Network for Saliency Prediction.【链接】
- Visual Saliency Detection Based on Multiscale Deep CNN Features.【链接】
- A Deep Spatial Contextual Long-term Recurrent Convolutional Network for Saliency Detection.【链接】
- Deeply supervised salient object detection with short connections.【链接】
- Weakly Supervised Top-down Salient Object Detection.【链接】
- SalGAN: Visual Saliency Prediction with Generative Adversarial Networks.【链接】
- Visual Saliency Prediction Using a Mixture of Deep Neural Networks.【链接】
- A Fast and Compact Salient Score Regression Network Based on Fully Convolutional Network.【链接】
- Saliency Detection by Forward and Backward Cues in Deep-CNNs.【链接】
- Supervised Adversarial Networks for Image Saliency Detection.【链接】
- Group-wise Deep Co-saliency Detection.【链接】
- Towards the Success Rate of One: Real-time Unconstrained Salient Object Detection.【链接】
- Amulet: Aggregating Multi-level Convolutional Features for Salient Object Detection.【链接】
- Learning Uncertain Convolutional Features for Accurate Saliency Detection.【链接】
- Deep Edge-Aware Saliency Detection.【链接】
- Self-explanatory Deep Salient Object Detection.【链接】
- PiCANet: Learning Pixel-wise Contextual Attention in ConvNets and Its Application in Saliency Detection.【链接】
- DeepFeat: A Bottom Up and Top Down Saliency Model Based on Deep Features of Convolutional Neural Nets.【链接】
Saliency Detection in Video
Visual Relationship Detection
- Visual Relationship Detection with Language Priors.【链接】
- ViP-CNN: A Visual Phrase Reasoning Convolutional Neural Network for Visual Relationship Detection.【链接】
- Visual Translation Embedding Network for Visual Relation Detection.【链接】
- Deep Variation-structured Reinforcement Learning for Visual Relationship and Attribute Detection.【链接】
- Detecting Visual Relationships with Deep Relational Networks.【链接】
- Identifying Spatial Relations in Images using Convolutional Neural Networks.【链接】
- PPR-FCN: Weakly Supervised Visual Relation Detection via Parallel Pairwise R-FCN.[【链接】][【链接】]【链接】
Specific Object Deteciton
- Deep Deformation Network for Object Landmark Localization.[【链接】]【链接】
- Fashion Landmark Detection in the Wild.【链接】
- Deep Learning for Fast and Accurate Fashion Item Detection.[【链接】]【链接】
- OSMDeepOD - OSM and Deep Learning based Object Detection from Aerial Imagery (formerly known as “OSM-Crosswalk-Detection”).【链接】
- Selfie Detection by Synergy-Constraint Based Convolutional Neural Network.【链接】
- Associative Embedding:End-to-End Learning for Joint Detection and Grouping.【链接】
- Deep Cuboid Detection: Beyond 2D Bounding Boxes.【链接】
- Automatic Model Based Dataset Generation for Fast and Accurate Crop and Weeds Detection.【链接】
- Deep Learning Logo Detection with Data Expansion by Synthesising Context.【链接】
- Pixel-wise Ear Detection with Convolutional Encoder-Decoder Networks.【链接】
- Automatic Handgun Detection Alarm in Videos Using Deep Learning.【链接】
- Using Deep Networks for Drone Detection.【链接】
- Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection.【链接】
- DeepVoting: An Explainable Framework for Semantic Part Detection under Partial Occlusion.【链接】
- Fast Shadow Detection from a Single Image Using a Patched Convolutional Neural Network.【链接】
Face Deteciton
- Multi-view Face Detection Using Deep Convolutional Neural Networks.【链接】
- From Facial Parts Responses to Face Detection: A Deep Learning Approach.【链接】
- Compact Convolutional Neural Network Cascade for Face Detection.【链接】
- Face Detection with End-to-End Integration of a ConvNet and a 3D Model.【链接】
- CMS-RCNN: Contextual Multi-Scale Region-based CNN for Unconstrained Face Detection.【链接】
- Finding Tiny Faces.【链接】
- Towards a Deep Learning Framework for Unconstrained Face Detection.【链接】
- Supervised Transformer Network for Efficient Face Detection.【链接】
UnitBox
MTCNN
- Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Neural Networks.【链接】
- Face Detection using Deep Learning: An Improved Faster RCNN Approach.【链接】
- Faceness-Net: Face Detection through Deep Facial Part Responses.【链接】
- Multi-Path Region-Based Convolutional Neural Network for Accurate Detection of Unconstrained “Hard Faces”.【链接】
- End-To-End Face Detection and Recognition.【链接】
- Face R-CNN.【链接】
- Face Detection through Scale-Friendly Deep Convolutional Networks.【链接】
- Scale-Aware Face Detection.【链接】
- Multi-Branch Fully Convolutional Network for Face Detection.【链接】
- SSH: Single Stage Headless Face Detector.【链接】
- Dockerface: an easy to install and use Faster R-CNN face detector in a Docker container.【链接】
- FaceBoxes: A CPU Real-time Face Detector with High Accuracy.【链接】
- S3FD: Single Shot Scale-invariant Face Detector.【链接】
- Detecting Faces Using Region-based Fully Convolutional Networks.【链接】
- AffordanceNet: An End-to-End Deep Learning Approach for Object Affordance Detection.【链接】
Facial Point / Landmark Detection
- Deep Convolutional Network Cascade for Facial Point Detection.【链接】
- Facial Landmark Detection by Deep Multi-task Learning.【链接】
- A Recurrent Encoder-Decoder Network for Sequential Face Alignment.【链接】
- Detecting facial landmarks in the video based on a hybrid framework.【链接】
- Deep Constrained Local Models for Facial Landmark Detection.(https://arxiv.org/abs/1611.08657)
- Effective face landmark localization via single deep network.【链接】
- A Convolution Tree with Deconvolution Branches: Exploiting Geometric Relationships for Single Shot Keypoint Detection.【链接】
- Deep Alignment Network: A convolutional neural network for robust face alignment.【链接】
- Joint Multi-view Face Alignment in the Wild.【链接】
- FacePoseNet: Making a Case for Landmark-Free Face Alignment.【链接】
People Detection
Person Head Detection
- Context-aware CNNs for person head detection.【链接】
Pedestrian Detection
- Pedestrian Detection aided by Deep Learning Semantic Tasks.【链接】
- Deep Learning Strong Parts for Pedestrian Detection.【链接】
- Taking a Deeper Look at Pedestrians.【链接】
- Convolutional Channel Features.【链接】
- Learning Complexity-Aware Cascades for Deep Pedestrian Detection.【链接】
- Deep convolutional neural networks for pedestrian detection.【链接】
- Scale-aware Fast R-CNN for Pedestrian Detection.【链接】
- New algorithm improves speed and accuracy of pedestrian detection.【链接】
- Pushing the Limits of Deep CNNs for Pedestrian Detection.【链接】
- A Real-Time Deep Learning Pedestrian Detector for Robot Navigation.【链接】
- A Real-Time Pedestrian Detector using Deep Learning for Human-Aware Navigation.【链接】
- Is Faster R-CNN Doing Well for Pedestrian Detection?【链接】
- Reduced Memory Region Based Deep Convolutional Neural Network Detection.【链接】
- Fused DNN: A deep neural network fusion approach to fast and robust pedestrian detection.【链接】
- Multispectral Deep Neural Networks for Pedestrian Detection.【链接】
- Expecting the Unexpected: Training Detectors for Unusual Pedestrians with Adversarial Imposters.【链接】
- Illuminating Pedestrians via Simultaneous Detection & Segmentation.【链接】
- Rotational Rectification Network for Robust Pedestrian Detection.【链接】
- STD-PD: Generating Synthetic Training Data for Pedestrian Detection in Unannotated Videos.【链接】
- Too Far to See? Not Really! — Pedestrian Detection with Scale-aware Localization Policy.【链接】
Vehicle Detection
Traffic-Sign Detection
Boundary / Edge / Contour Detection
- Holistically-Nested Edge Detection.【链接】
- Unsupervised Learning of Edges.【链接】
- Pushing the Boundaries of Boundary Detection using Deep Learning.【链接】
- Convolutional Oriented Boundaries.【链接】
- Convolutional Oriented Boundaries: From Image Segmentation to High-Level Tasks.【链接】
- Richer Convolutional Features for Edge Detection.【链接】
- Contour Detection from Deep Patch-level Boundary Prediction.【链接】
- CASENet: Deep Category-Aware Semantic Edge Detection.【链接】
Skeleton Detection
Fruit Detection
Part Detection
- Objects as context for part detection.【链接】
Object Proposa
- DeepProposal: Hunting Objects by Cascading Deep Convolutional Layers.【链接】
- Scale-aware Pixel-wise Object Proposal Networks.【链接】
- Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization.【链接】
- Learning to Segment Object Proposals via Recursive Neural Networks.【链接】
- Learning Detection with Diverse Proposals.【链接】
- ScaleNet: Guiding Object Proposal Generation in Supermarkets and Beyond.【链接】
- Improving Small Object Proposals for Company Logo Detection.【链接】
Localization
- Beyond Bounding Boxes: Precise Localization of Objects in Images.【链接】
- Weakly Supervised Object Localization with Multi-fold Multiple Instance Learning.【链接】
- Weakly Supervised Object Localization Using Size Estimates.【链接】
- Active Object Localization with Deep Reinforcement Learning.【链接】
- Localizing objects using referring expressions.【链接】
- LocNet: Improving Localization Accuracy for Object Detection.【链接】
- Learning Deep Features for Discriminative Localization.【链接】
- ContextLocNet: Context-Aware Deep Network Models for Weakly Supervised Localization.【链接】
- Ensemble of Part Detectors for Simultaneous Classification and Localization.【链接】
- STNet: Selective Tuning of Convolutional Networks for Object Localization.【链接】
- Soft Proposal Networks for Weakly Supervised Object Localization.【链接】
- Fine-grained Discriminative Localization via Saliency-guided Faster R-CNN.【链接】
Tutorials / Talks
Projects
- TensorBox: a simple framework for training neural networks to detect objects in images.【链接】
- Object detection in torch: Implementation of some object detection frameworks in torch.【链接】
- Using DIGITS to train an Object Detection network.【链接】
- FCN-MultiBox Detector.【链接】
- KittiBox: A car detection model implemented in Tensorflow.【链接】
- Deformable Convolutional Networks + MST + Soft-NMS.【链接】
Tools
- BeaverDam: Video annotation tool for deep learning training labels.【链接】
Blogs
- Convolutional Neural Networks for Object Detection.【链接】
- Introducing automatic object detection to visual search (Pinterest).【链接】
- Deep Learning for Object Detection with DIGITS.【链接】
- Analyzing The Papers Behind Facebook’s Computer Vision Approach.【链接】
- Easily Create High Quality Object Detectors with Deep Learning.【链接】
- How to Train a Deep-Learned Object Detection Model in the Microsoft Cognitive Toolkit.【链接】
- Object Detection in Satellite Imagery, a Low Overhead Approach.【链接】
- You Only Look Twice — Multi-Scale Object Detection in Satellite Imagery With Convolutional Neural Networks.【链接】
- Faster R-CNN Pedestrian and Car Detection.【链接】
- Small U-Net for vehicle detection.【链接】
- Region of interest pooling explained.【链接】
- Supercharge your Computer Vision models with the TensorFlow Object Detection API.【链接】
(作者:朱坤)