目标检测领域 2015-2

Shallow and Deep Convolutional Networks for Saliency Prediction

arxiv:http://arxiv.org/abs/1603.00845

github:https://github.com/imatge-upc/saliency-2016-cvpr

Recurrent Attentional Networks for Saliency Detection

intro: CVPR 2016. recurrent attentional convolutional-deconvolution network (RACDNN)

arxiv:http://arxiv.org/abs/1604.03227

Two-Stream Convolutional Networks for Dynamic Saliency Prediction

arxiv:http://arxiv.org/abs/1607.04730

Unconstrained Salient Object Detection

Unconstrained Salient Object Detection via Proposal Subset Optimization

intro: CVPR 2016

project page:http://cs-people.bu.edu/jmzhang/sod.html

paper:http://cs-people.bu.edu/jmzhang/SOD/CVPR16SOD_camera_ready.pdf

github:https://github.com/jimmie33/SOD

caffe model zoo:https://github.com/BVLC/caffe/wiki/Model-Zoo#cnn-object-proposal-models-for-salient-object-detection

DHSNet: Deep Hierarchical Saliency Network for Salient Object Detection

paper:http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Liu_DHSNet_Deep_Hierarchical_CVPR_2016_paper.pdf

Salient Object Subitizing

intro: CVPR 2015

intro: predicting the existence and the number of salient objects in an image using holistic cues

project page:http://cs-people.bu.edu/jmzhang/sos.html

arxiv:http://arxiv.org/abs/1607.07525

paper:http://cs-people.bu.edu/jmzhang/SOS/SOS_preprint.pdf

caffe model zoo:https://github.com/BVLC/caffe/wiki/Model-Zoo#cnn-models-for-salient-object-subitizing

Deeply-Supervised Recurrent Convolutional Neural Network for Saliency Detection

intro: ACMMM 2016. deeply-supervised recurrent convolutional neural network (DSRCNN)

arxiv:http://arxiv.org/abs/1608.05177

Saliency Detection via Combining Region-Level and Pixel-Level Predictions with CNNs

intro: ECCV 2016

arxiv:http://arxiv.org/abs/1608.05186

Edge Preserving and Multi-Scale Contextual Neural Network for Salient Object Detection

arxiv:http://arxiv.org/abs/1608.08029

A Deep Multi-Level Network for Saliency Prediction

arxiv:http://arxiv.org/abs/1609.01064

Visual Saliency Detection Based on Multiscale Deep CNN Features

intro: IEEE Transactions on Image Processing

arxiv:http://arxiv.org/abs/1609.02077

A Deep Spatial Contextual Long-term Recurrent Convolutional Network for Saliency Detection

intro: DSCLRCN

arxiv:https://arxiv.org/abs/1610.01708

Deeply supervised salient object detection with short connections

arxiv:https://arxiv.org/abs/1611.04849

Weakly Supervised Top-down Salient Object Detection

intro: Nanyang Technological University

arxiv:https://arxiv.org/abs/1611.05345

Specific Object Deteciton

Face Deteciton

Multi-view Face Detection Using Deep Convolutional Neural Networks

intro: Yahoo

arxiv:http://arxiv.org/abs/1502.02766

From Facial Parts Responses to Face Detection: A Deep Learning Approach

project page:https://kpzhang93.github.io/MTCNN_face_detection_alignment/index.html

arxiv:https://arxiv.org/abs/1604.02878

github(Matlab):https://github.com/kpzhang93/MTCNN_face_detection_alignment

github(MXNet):https://github.com/pangyupo/mxnet_mtcnn_face_detection

github:https://github.com/DaFuCoding/MTCNN_Caffe

Datasets / Benchmarks

FDDB: Face Detection Data Set and Benchmark

homepage:http://vis-www.cs.umass.edu/fddb/index.html

results:http://vis-www.cs.umass.edu/fddb/results.html

WIDER FACE: A Face Detection Benchmark

homepage:http://mmlab.ie.cuhk.edu.hk/projects/WIDERFace/

arxiv:http://arxiv.org/abs/1511.06523

Facial Point / Landmark Detection

Deep Convolutional Network Cascade for Facial Point Detection

homepage:http://mmlab.ie.cuhk.edu.hk/archive/CNN_FacePoint.htm

paper:http://www.ee.cuhk.edu.hk/~xgwang/papers/sunWTcvpr13.pdf

github:https://github.com/luoyetx/deep-landmark

A Recurrent Encoder-Decoder Network for Sequential Face Alignment

intro: ECCV 2016

arxiv:https://arxiv.org/abs/1608.05477

Detecting facial landmarks in the video based on a hybrid framework

arxiv:http://arxiv.org/abs/1609.06441

Deep Constrained Local Models for Facial Landmark Detection

arxiv:https://arxiv.org/abs/1611.08657

People Detection

End-to-end people detection in crowded scenes

arxiv:http://arxiv.org/abs/1506.04878

github:https://github.com/Russell91/reinspect

ipn:http://nbviewer.ipython.org/github/Russell91/ReInspect/blob/master/evaluation_reinspect.ipynb

Detecting People in Artwork with CNNs

intro: ECCV 2016 Workshops

arxiv:https://arxiv.org/abs/1610.08871

Person Head Detection

Context-aware CNNs for person head detection

arxiv:http://arxiv.org/abs/1511.07917

github:https://github.com/aosokin/cnn_head_detection

Pedestrian Detection

Pedestrian Detection aided by Deep Learning Semantic Tasks

intro: CVPR 2015

project page:http://mmlab.ie.cuhk.edu.hk/projects/TA-CNN/

paper:http://arxiv.org/abs/1412.0069

Deep Learning Strong Parts for Pedestrian Detection

intro: ICCV 2015. CUHK. DeepParts

intro: Achieving 11.89% average miss rate on Caltech Pedestrian Dataset

paper:http://personal.ie.cuhk.edu.hk/~pluo/pdf/tianLWTiccv15.pdf

Deep convolutional neural networks for pedestrian detection

arxiv:http://arxiv.org/abs/1510.03608

github:https://github.com/DenisTome/DeepPed

New algorithm improves speed and accuracy of pedestrian detection

blog:http://www.eurekalert.org/pub_releases/2016-02/uoc–nai020516.php

Pushing the Limits of Deep CNNs for Pedestrian Detection

intro: “set a new record on the Caltech pedestrian dataset, lowering the log-average miss rate from 11.7% to 8.9%”

arxiv:http://arxiv.org/abs/1603.04525

A Real-Time Deep Learning Pedestrian Detector for Robot Navigation

arxiv:http://arxiv.org/abs/1607.04436

A Real-Time Pedestrian Detector using Deep Learning for Human-Aware Navigation

arxiv:http://arxiv.org/abs/1607.04441

Is Faster R-CNN Doing Well for Pedestrian Detection?

arxiv:http://arxiv.org/abs/1607.07032

github:https://github.com/zhangliliang/RPN_BF/tree/RPN-pedestrian

Reduced Memory Region Based Deep Convolutional Neural Network Detection

intro: IEEE 2016 ICCE-Berlin

arxiv:http://arxiv.org/abs/1609.02500

Fused DNN: A deep neural network fusion approach to fast and robust pedestrian detection

arxiv:https://arxiv.org/abs/1610.03466

Multispectral Deep Neural Networks for Pedestrian Detection

intro: BMVC 2016 oral

arxiv:https://arxiv.org/abs/1611.02644

Vehicle Detection

DAVE: A Unified Framework for Fast Vehicle Detection and Annotation

intro: ECCV 2016

arxiv:http://arxiv.org/abs/1607.04564

Traffic-Sign Detection

Traffic-Sign Detection and Classification in the Wild

project page(code+dataset):http://cg.cs.tsinghua.edu.cn/traffic-sign/

paper:http://120.52.73.11/www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Zhu_Traffic-Sign_Detection_and_CVPR_2016_paper.pdf

code & model:http://cg.cs.tsinghua.edu.cn/traffic-sign/data_model_code/newdata0411.zip

Boundary / Edge / Contour Detection

Holistically-Nested Edge Detection

intro: ICCV 2015, Marr Prize

paper:http://www.cv-foundation.org/openaccess/content_iccv_2015/papers/Xie_Holistically-Nested_Edge_Detection_ICCV_2015_paper.pdf

arxiv:http://arxiv.org/abs/1504.06375

github:https://github.com/s9xie/hed

Unsupervised Learning of Edges

intro: CVPR 2016. Facebook AI Research

arxiv:http://arxiv.org/abs/1511.04166

zn-blog:http://www.leiphone.com/news/201607/b1trsg9j6GSMnjOP.html

Pushing the Boundaries of Boundary Detection using Deep Learning

arxiv:http://arxiv.org/abs/1511.07386

Convolutional Oriented Boundaries

intro: ECCV 2016

arxiv:http://arxiv.org/abs/1608.02755

Richer Convolutional Features for Edge Detection

intro: richer convolutional features (RCF)

arxiv:https://arxiv.org/abs/1612.02103

Skeleton Detection

Object Skeleton Extraction in Natural Images by Fusing Scale-associated Deep Side Outputs

arxiv:http://arxiv.org/abs/1603.09446

github:https://github.com/zeakey/DeepSkeleton

DeepSkeleton: Learning Multi-task Scale-associated Deep Side Outputs for Object Skeleton Extraction in Natural Images

arxiv:http://arxiv.org/abs/1609.03659

Fruit Detection

Deep Fruit Detection in Orchards

arxiv:https://arxiv.org/abs/1610.03677

Image Segmentation for Fruit Detection and Yield Estimation in Apple Orchards

intro: The Journal of Field Robotics in May 2016

project page:http://confluence.acfr.usyd.edu.au/display/AGPub/

arxiv:https://arxiv.org/abs/1610.08120

Others

Deep Deformation Network for Object Landmark Localization

arxiv:http://arxiv.org/abs/1605.01014

Fashion Landmark Detection in the Wild

arxiv:http://arxiv.org/abs/1608.03049

Deep Learning for Fast and Accurate Fashion Item Detection

intro: Kuznech Inc.

intro: MultiBox and Fast R-CNN

paper:https://kddfashion2016.mybluemix.net/kddfashion_finalSubmissions/Deep%20Learning%20for%20Fast%20and%20Accurate%20Fashion%20Item%20Detection.pdf

Visual Relationship Detection with Language Priors

intro: ECCV 2016 oral

paper:https://cs.stanford.edu/people/ranjaykrishna/vrd/vrd.pdf

github:https://github.com/Prof-Lu-Cewu/Visual-Relationship-Detection

OSMDeepOD - OSM and Deep Learning based Object Detection from Aerial Imagery (formerly known as “OSM-Crosswalk-Detection”)

github:https://github.com/geometalab/OSMDeepOD

Selfie Detection by Synergy-Constraint Based Convolutional Neural Network

intro: IEEE SITIS 2016

arxiv:https://arxiv.org/abs/1611.04357

Associative Embedding:End-to-End Learning for Joint Detection and Grouping

arxiv:https://arxiv.org/abs/1611.05424

Deep Cuboid Detection: Beyond 2D Bounding Boxes

intro: CMU & Magic Leap

arxiv:https://arxiv.org/abs/1611.10010

Object Proposal

DeepProposal: Hunting Objects by Cascading Deep Convolutional Layers

arxiv:http://arxiv.org/abs/1510.04445

github:https://github.com/aghodrati/deepproposal

Scale-aware Pixel-wise Object Proposal Networks

intro: IEEE Transactions on Image Processing

arxiv:http://arxiv.org/abs/1601.04798

Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization

intro: AttractioNet

arxiv:https://arxiv.org/abs/1606.04446

github:https://github.com/gidariss/AttractioNet

Learning to Segment Object Proposals via Recursive Neural Networks

arxiv:https://arxiv.org/abs/1612.01057

Localization

Beyond Bounding Boxes: Precise Localization of Objects in Images

intro: PhD Thesis

homepage:http://www.eecs.berkeley.edu/Pubs/TechRpts/2015/EECS-2015-193.html

phd-thesis:http://www.eecs.berkeley.edu/Pubs/TechRpts/2015/EECS-2015-193.pdf

github(“SDS using hypercolumns”):https://github.com/bharath272/sds

Weakly Supervised Object Localization with Multi-fold Multiple Instance Learning

arxiv:http://arxiv.org/abs/1503.00949

Weakly Supervised Object Localization Using Size Estimates

arxiv:http://arxiv.org/abs/1608.04314

Localizing objects using referring expressions

intro: ECCV 2016

keywords: LSTM, multiple instance learning (MIL)

paper:http://www.umiacs.umd.edu/~varun/files/refexp-ECCV16.pdf

github:https://github.com/varun-nagaraja/referring-expressions

LocNet: Improving Localization Accuracy for Object Detection

arxiv:http://arxiv.org/abs/1511.07763

github:https://github.com/gidariss/LocNet

Learning Deep Features for Discriminative Localization

homepage:http://cnnlocalization.csail.mit.edu/

arxiv:http://arxiv.org/abs/1512.04150

github(Tensorflow):https://github.com/jazzsaxmafia/Weakly_detector

github:https://github.com/metalbubble/CAM

github:https://github.com/tdeboissiere/VGG16CAM-keras

ContextLocNet: Context-Aware Deep Network Models for Weakly Supervised Localization

intro: ECCV 2016

project page:http://www.di.ens.fr/willow/research/contextlocnet/

arxiv:http://arxiv.org/abs/1609.04331

github:https://github.com/vadimkantorov/contextlocnet

Tutorials

Convolutional Feature Maps: Elements of efficient (and accurate) CNN-based object detection

slides:http://research.microsoft.com/en-us/um/people/kahe/iccv15tutorial/iccv2015_tutorial_convolutional_feature_maps_kaiminghe.pdf

Projects

TensorBox: a simple framework for training neural networks to detect objects in images

intro: “The basic model implements the simple and robust GoogLeNet-OverFeat algorithm. We additionally provide an implementation of theReInspectalgorithm”

github:https://github.com/Russell91/TensorBox

Object detection in torch: Implementation of some object detection frameworks in torch

github:https://github.com/fmassa/object-detection.torch

Using DIGITS to train an Object Detection network

github:https://github.com/NVIDIA/DIGITS/blob/master/examples/object-detection/README.md

FCN-MultiBox Detector

intro: Full convolution MultiBox Detector ( like SSD) implemented in Torch.

github:https://github.com/teaonly/FMD.torch

Blogs

Convolutional Neural Networks for Object Detection

http://rnd.azoft.com/convolutional-neural-networks-object-detection/

Introducing automatic object detection to visual search (Pinterest)

keywords: Faster R-CNN

blog:https://engineering.pinterest.com/blog/introducing-automatic-object-detection-visual-search

demo:https://engineering.pinterest.com/sites/engineering/files/Visual%20Search%20V1%20-%20Video.mp4

review:https://news.developer.nvidia.com/pinterest-introduces-the-future-of-visual-search/?mkt_tok=eyJpIjoiTnpaa01UWXpPRE0xTURFMiIsInQiOiJJRjcybjkwTmtmallORUhLOFFFODBDclFqUlB3SWlRVXJXb1MrQ013TDRIMGxLQWlBczFIeWg0TFRUdnN2UHY2ZWFiXC9QQVwvQzBHM3B0UzBZblpOSmUyU1FcLzNPWXI4cml2VERwTTJsOFwvOEk9In0%3D

Deep Learning for Object Detection with DIGITS

blog:https://devblogs.nvidia.com/parallelforall/deep-learning-object-detection-digits/

Analyzing The Papers Behind Facebook’s Computer Vision Approach

keywords: DeepMask, SharpMask, MultiPathNet

blog:https://adeshpande3.github.io/adeshpande3.github.io/Analyzing-the-Papers-Behind-Facebook’s-Computer-Vision-Approach/

**Easily Create High Quality Object Detectors with Deep Learning **

intro: dlib v19.2

blog:http://blog.dlib.net/2016/10/easily-create-high-quality-object.html

How to Train a Deep-Learned Object Detection Model in the Microsoft Cognitive Toolkit

blog:https://blogs.technet.microsoft.com/machinelearning/2016/10/25/how-to-train-a-deep-learned-object-detection-model-in-cntk/

github:https://github.com/Microsoft/CNTK/tree/master/Examples/Image/Detection/FastRCNN

Object Detection in Satellite Imagery, a Low Overhead Approach

part 1:https://medium.com/the-downlinq/object-detection-in-satellite-imagery-a-low-overhead-approach-part-i-cbd96154a1b7#.2csh4iwx9

part 2:https://medium.com/the-downlinq/object-detection-in-satellite-imagery-a-low-overhead-approach-part-ii-893f40122f92#.f9b7dgf64

You Only Look Twice — Multi-Scale Object Detection in Satellite Imagery With Convolutional Neural Networks

part 1:https://medium.com/the-downlinq/you-only-look-twice-multi-scale-object-detection-in-satellite-imagery-with-convolutional-neural-38dad1cf7571#.fmmi2o3of

part 2:https://medium.com/the-downlinq/you-only-look-twice-multi-scale-object-detection-in-satellite-imagery-with-convolutional-neural-34f72f659588#.nwzarsz1t

Faster R-CNN Pedestrian and Car Detection

blog:https://bigsnarf.wordpress.com/2016/11/07/faster-r-cnn-pedestrian-and-car-detection/

ipn:https://gist.github.com/bigsnarfdude/2f7b2144065f6056892a98495644d3e0#file-demo_faster_rcnn_notebook-ipynb

github:https://github.com/bigsnarfdude/Faster-RCNN_TF