Object Detection. TA : Young-geun Kim. Biostatistics Lab., Seoul National University. March-June, 2018

Similar documents
Object Detection Based on Deep Learning

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Spatial Localization and Detection. Lecture 8-1

Object detection with CNNs

Unified, real-time object detection

Fine-tuning Pre-trained Large Scaled ImageNet model on smaller dataset for Detection task

Object detection using Region Proposals (RCNN) Ernest Cheung COMP Presentation

Lecture 5: Object Detection

Object Detection. CS698N Final Project Presentation AKSHAT AGARWAL SIDDHARTH TANWAR

Yiqi Yan. May 10, 2017

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

YOLO9000: Better, Faster, Stronger

Deep learning for object detection. Slides from Svetlana Lazebnik and many others

Optimizing Object Detection:

Final Report: Smart Trash Net: Waste Localization and Classification

Extend the shallow part of Single Shot MultiBox Detector via Convolutional Neural Network

CS6501: Deep Learning for Visual Recognition. Object Detection I: RCNN, Fast-RCNN, Faster-RCNN

REGION AVERAGE POOLING FOR CONTEXT-AWARE OBJECT DETECTION

Rich feature hierarchies for accurate object detection and semantic segmentation

Object Detection on Self-Driving Cars in China. Lingyun Li

arxiv: v1 [cs.cv] 4 Jun 2015

G-CNN: an Iterative Grid Based Object Detector

Project 3 Q&A. Jonathan Krause

OBJECT DETECTION HYUNG IL KOO

Deep Learning for Object detection & localization

Direct Multi-Scale Dual-Stream Network for Pedestrian Detection Sang-Il Jung and Ki-Sang Hong Image Information Processing Lab.

Optimizing Intersection-Over-Union in Deep Neural Networks for Image Segmentation

Introduction to Deep Learning for Facial Understanding Part III: Regional CNNs

Category-level localization

3 Object Detection. BVM 2018 Tutorial: Advanced Deep Learning Methods. Paul F. Jaeger, Division of Medical Image Computing

Gradient of the lower bound

Rich feature hierarchies for accurate object detection and semantic segmentation

CS 1674: Intro to Computer Vision. Object Recognition. Prof. Adriana Kovashka University of Pittsburgh April 3, 5, 2018

YOLO: You Only Look Once Unified Real-Time Object Detection. Presenter: Liyang Zhong Quan Zou

Object Detection with YOLO on Artwork Dataset

Deformable Part Models

Object Recognition II

Automatic detection of books based on Faster R-CNN

CPSC340. State-of-the-art Neural Networks. Nando de Freitas November, 2012 University of British Columbia

Feature-Fused SSD: Fast Detection for Small Objects

Efficient Segmentation-Aided Text Detection For Intelligent Robots

Optimizing Object Detection:

Classification of objects from Video Data (Group 30)

Detection and Localization with Multi-scale Models

Deep Learning in Visual Recognition. Thanks Da Zhang for the slides

Rich feature hierarchies for accurate object detection and semant

R-FCN++: Towards Accurate Region-Based Fully Convolutional Networks for Object Detection

Real-time Object Detection CS 229 Course Project

MULTI-SCALE OBJECT DETECTION WITH FEATURE FUSION AND REGION OBJECTNESS NETWORK. Wenjie Guan, YueXian Zou*, Xiaoqun Zhou

An Object Detection Algorithm based on Deformable Part Models with Bing Features Chunwei Li1, a and Youjun Bu1, b

CIS680: Vision & Learning Assignment 2.b: RPN, Faster R-CNN and Mask R-CNN Due: Nov. 21, 2018 at 11:59 pm

Modern Convolutional Object Detectors

Traffic Multiple Target Detection on YOLOv2

arxiv: v1 [cs.cv] 3 Apr 2016

Mask R-CNN. presented by Jiageng Zhang, Jingyao Zhan, Yunhan Ma

Towards Real-Time Automatic Number Plate. Detection: Dots in the Search Space

Cascade Region Regression for Robust Object Detection

PT-NET: IMPROVE OBJECT AND FACE DETECTION VIA A PRE-TRAINED CNN MODEL

Paper Motivation. Fixed geometric structures of CNN models. CNNs are inherently limited to model geometric transformations

Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks

Object Detection in Sports Videos

Regionlet Object Detector with Hand-crafted and CNN Feature

International Journal of Computer Engineering and Applications, Volume XII, Special Issue, September 18,

Hand Detection For Grab-and-Go Groceries

A Novel Representation and Pipeline for Object Detection

Yield Estimation using faster R-CNN

Exploiting Depth from Single Monocular Images for Object Detection and Semantic Segmentation

SSD: Single Shot MultiBox Detector

arxiv: v1 [cs.cv] 15 Oct 2018

Supplementary Material: Pixelwise Instance Segmentation with a Dynamically Instantiated Network

Computer Vision Lecture 16

arxiv: v1 [cs.cv] 26 May 2017

[Supplementary Material] Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors

Finding Tiny Faces Supplementary Materials

Visual features detection based on deep neural network in autonomous driving tasks

Beyond Sliding Windows: Object Localization by Efficient Subwindow Search

arxiv: v1 [cs.cv] 15 Aug 2018

PASCAL VOC Classification: Local Features vs. Deep Features. Shuicheng YAN, NUS

R-FCN: OBJECT DETECTION VIA REGION-BASED FULLY CONVOLUTIONAL NETWORKS

Single-Shot Refinement Neural Network for Object Detection -Supplementary Material-

Object Detection and Its Implementation on Android Devices

CNN BASED REGION PROPOSALS FOR EFFICIENT OBJECT DETECTION. Jawadul H. Bappy and Amit K. Roy-Chowdhury

Deep condolence to Professor Mark Everingham

Computer Vision Lecture 16

AttentionNet for Accurate Localization and Detection of Objects. (To appear in ICCV 2015)

MCMOT: Multi-Class Multi-Object Tracking using Changing Point Detection

Mask R-CNN. Kaiming He, Georgia, Gkioxari, Piotr Dollar, Ross Girshick Presenters: Xiaokang Wang, Mengyao Shi Feb. 13, 2018

Future directions in computer vision. Larry Davis Computer Vision Laboratory University of Maryland College Park MD USA

HIERARCHICAL JOINT-GUIDED NETWORKS FOR SEMANTIC IMAGE SEGMENTATION

Volume 6, Issue 12, December 2018 International Journal of Advance Research in Computer Science and Management Studies

arxiv: v3 [cs.cv] 15 Sep 2018

Computer Vision Lecture 16

Industrial Technology Research Institute, Hsinchu, Taiwan, R.O.C ǂ

Learning Detection with Diverse Proposals

SSD: Single Shot MultiBox Detector. Author: Wei Liu et al. Presenter: Siyu Jiang

Mimicking Very Efficient Network for Object Detection

Content-Based Image Recovery

Subcategory-aware Convolutional Neural Networks for Object Proposals and Detection

Rotation Invariance Neural Network

Hierarchical Image-Region Labeling via Structured Learning

Transcription:

Object Detection TA : Young-geun Kim Biostatistics Lab., Seoul National University March-June, 2018 Seoul National University Deep Learning March-June, 2018 1 / 57

Index 1 Introduction 2 R-CNN 3 YOLO 4 Evaluation Seoul National University Deep Learning March-June, 2018 2 / 57

Introduction Introduction Seoul National University Deep Learning March-June, 2018 3 / 57

Introduction In this session, we will learn about... The Object Detection problem. Regions with CNN features (R-CNN), a region proposal-based approach model. You-Only-Look-Once (YOLO), an unified approach model. Evaluation metrics for detection models. Seoul National University Deep Learning March-June, 2018 4 / 57

Introduction What is Object Detection? Object Detection is a task finding where and what objects are (Location + Classification). An integral part of various vision application such as Automated Driving System, Face Detection and Object Counting. Figure: from https://youtu.be/mpu2histivi (YOLO v3 clip). Seoul National University Deep Learning March-June, 2018 5 / 57

Introduction What is Object Detection? (Conti.) For given image, task-taker should answer the predicted region and class confidence. A region is expressed as rectangular called bounding box. The number of objects is not provided. Figure: from Ren et al., 2015.. Seoul National University Deep Learning March-June, 2018 6 / 57

Introduction What is Object Detection? (Conti.) Exact region (or bounding box) of each object is called Ground-Truth (GT) box, the minimal rectangular containing whole part of the object. A region is parameterized by (x, y, w, h) where (x, y) is the coordinate of top-left (or center) point, w is the width, and h is the height of the bounding box. Intersection over Union (IoU), the ratio of intersection area to union area, between predicted region and GT box presents the accuracy about location. Seoul National University Deep Learning March-June, 2018 7 / 57

Introduction What is Object Detection? (Conti.) There are various types of objects. For example, VOC challenge requires detecting following 20 kinds of object classes. Middle Level Person Animal Vehicle Indoor Low Level person bird, cat, cow, dog, horse, sheep aeroplane, bicycle, boat, bus, car, motorbike, train bottle, chair, diningtable, potted plant, sofa, tv/monitor mean Average Precision (map), an estimator of the area under the precision-recall curve (AUCPR), usually presents the accuracy about classification. Seoul National University Deep Learning March-June, 2018 8 / 57

Introduction What is Object Detection? (Conti.) An object is considered detected if task-taker selects any region with predicted label satisfying following conditions. Condition 1 : Highly overlapped with GT box of the object. Condition 2 : Correctly classified. Seoul National University Deep Learning March-June, 2018 9 / 57

Introduction Challenges Infinitely Imbalanced Structure : Background (BG) is the majority class. There are few positive regions (GT) and infinitely many negative regions (BG). True Predicted N P Total N TN FP # of BG P FN TP # of GT Table: The confusion matrix of object detection. In this structure, accuracy about positive class is severely suffered. This means that finding an object position as it is difficult. Seoul National University Deep Learning March-June, 2018 10 / 57

Introduction Challenges (Conti.) Dynamic Scale : Shape of objects is various. Some are tiny/huge and some are horizontally/vertically long. Figure: from VOC2012. This means that our model should recognize various scale of regions. Seoul National University Deep Learning March-June, 2018 11 / 57

Introduction Challenges (Conti.) Multi-task : Finding object position (Location) and classifying the object (Classification) each is difficult. Object Detection requires performing both tasks simultaneously. In practice, the test time of detection model should be short, but due to the high level of difficulty, it is challenging. Seoul National University Deep Learning March-June, 2018 12 / 57

Introduction Approaches Pre-deep learning approaches (Do not cover. See 50 years of object recognition: Directions forward). Regions with CNN features (R-CNN), a region proposal-based approach model. You-Only-Look-Once (YOLO), an unified approach model. Seoul National University Deep Learning March-June, 2018 13 / 57

R-CNN R-CNN Seoul National University Deep Learning March-June, 2018 14 / 57

R-CNN Regions with CNN features Regions with CNN features (R-CNN) is a region proposal-based approach model (Girshick et el., 2014). R-CNN selects regions using Selective Search (Uijlinga et al., 2013), warps them as to the same scale and extracts features to learn class-specific SVMs. Figure: from Girshick et el., 2014. Seoul National University Deep Learning March-June, 2018 15 / 57

R-CNN Selective Search Selective Search (SS) is an hierarchical grouping algorithm whose domain is a set of region. For given set of regions, SS greedily merges regions. The distance measure is a partial summation of similarity about colour, texture, size, and fill. Figure: from Uijlinga et al., 2013. Seoul National University Deep Learning March-June, 2018 16 / 57

R-CNN Selective Search (Conti.) Considering various features from fine-level region, SS distinguishes objects and captures their hierarchical structure. Initialization is based on a graph-based segmentation algorithm (Felzenszwalb and Huttenlocher, 2004.) whose time complexity is nearly linear in the number of pixels. Seoul National University Deep Learning March-June, 2018 17 / 57

R-CNN Detection Network Proposed regions pass through CNNs which consists of classifier and bounding box (bbox) regressor. AlexNet (Krizhevsky et al., 2012.) is applied with replaced FC layer for corresponding number of class including BG. After tuning AlexNet, class-specific linear SVMs and bbox regressor (Felzenszwalb et al., 2010.) are learned by using extracted feature. bbox regressor predicts (x, y, w, h) of GT and use it to adjust proposed regions. Seoul National University Deep Learning March-June, 2018 18 / 57

R-CNN Limitation of R-CNN R-CNN requires fine-tuning CNN, learning multiple SVMs and bbox regressor (multi-stage pipeline). Training SVMs and bbox regressor requires feedforwarding all regions in all images and saving all extracted features. Because of the same reason in training, test time is too long. It takes 47 second to perform detection for a single image. Seoul National University Deep Learning March-June, 2018 19 / 57

R-CNN Spatial Pyramid Pooling Network Feedforwarding all proposed regions is time-consuming approach. Spatial Pyramid Pooling (SPP; He et al., 2014.) models the spatial connectivity and makes various regions into the fixed size. For usual CNNs, warp conv conv warp since there is no spatial connectivity between raw image and extracted feature. Figure: from He et al., 2014. Seoul National University Deep Learning March-June, 2018 20 / 57

R-CNN Spatial Pyramid Pooling Network (Conti.) SPP Network learns the spatial connectivity, still capturing semantic content. SPP reduces the computation cost, but SPP Network is still multi-stage pipeline. Figure: from He et al., 2014. Seoul National University Deep Learning March-June, 2018 21 / 57

R-CNN Fast R-CNN Fast R-CNN (Girshick and Ross, 2015.) is a variation of R-CNN applying Region of Interest (RoI) pooling, a kind of SPP. Training is single-stage by using multi-task loss. Multi-task loss enables us to update all weights simultaneously. Figure: from Girshick and Ross, 2015. Seoul National University Deep Learning March-June, 2018 22 / 57

R-CNN Region of Interest Pooling RoI pooling connects the raw image and the final extracted feature before FC layers. RoI feature vector passes two sibling FC layer. Figure: from Girshick and Ross, 2015. Seoul National University Deep Learning March-June, 2018 23 / 57

R-CNN Region of Interest Pooling (Conti.) In contrast to usual max-pooling, RoI pooling has dynamic filter size. Back propagation through RoI pooling requires activated positions for each region. Seoul National University Deep Learning March-June, 2018 24 / 57

R-CNN Multi-task Loss For given region (x r, y r, w r, h r ) in an image, Fast R-CNN calculates p and t k = (tx k, ty k, tw k, th k ), the predicted probability vector and location for class k parameterized by following. t k x = (x k x r )/w r t k y = (y k y r )/h r t k w = log(w k /w r ) t k h = log(hk /h r ) Let u, v be the true class and location of corresponding GT box for given region. v is parameterized by substituting (x, y, w, h) of the GT. Seoul National University Deep Learning March-June, 2018 25 / 57

R-CNN Multi-task Loss (Conti.) To train multi-task model, the loss function is designed as L(p, u, t u, v) = L cls (p, u) + λ[u 1]L loc (t u, v) where L cls (p, u) = log p u is log loss for true class u and L loc (t u, v) = i {x,y,w,h} huber(t u i v i ). The hyper-parameter λ controls balance between classification loss and regression loss. For u = 0, background region, L loc doesn t have any role. L loc is a function of ti u v i, so L is invariant to translation, flipping and rescaling. Seoul National University Deep Learning March-June, 2018 26 / 57

R-CNN Limitation of Fast R-CNN (Conti.) Compared to R-CNN, Fast R-CNN achieves slightly higher accuracy with nearly 100 times short test time, but the test time is still long. In VOC 2007 test task, Fast R-CNN takes 1830ms per image. Region proposal task, SS is a huge piece consuming 1510ms per image. Seoul National University Deep Learning March-June, 2018 27 / 57

R-CNN Faster R-CNN Faster R-CNN (Ren et al., 2015) is a variation of Fast R-CNN using Region Proposal Network (RPN). Roughly speaking, Faster R-CNN = RPN + Fast R-CNN. Contrast to SS, RPN has learnable weight for multi-task loss. Seoul National University Deep Learning March-June, 2018 28 / 57

R-CNN Region Proposal Network For given point in an image, RPN classifies objectness of several regions centered on that point and regresses exact location. Pre-determined points in each image are called anchors. Figure: adjusted from VOC2012. Seoul National University Deep Learning March-June, 2018 29 / 57

R-CNN Region Proposal Network (Conti.) 1. For selected anchor, view small region nearby the anchor in the level of extracted feature. 2. Determine the objectness of k regions centered on the corresponding anchor in the raw image map. Figure: from Ren et al., 2015. Seoul National University Deep Learning March-June, 2018 30 / 57

R-CNN Region Proposal Network (Conti.) 3. For all regions classified to be positive, adjust them using reg layer. Figure: from Ren et al., 2015. Seoul National University Deep Learning March-June, 2018 31 / 57

R-CNN Multi-task loss for RPN RPN uses multi-task loss similar to Fast R-CNN. Exact formula is L({p i }, {t i }) = 1 L cls (p i, pi ) + λ 1 pi L reg (t i, ti ) N cls N reg where i is the index of an anchor. i Here, p i is the predicted probability of anchor i being an object. pi the ground-truth label. t is the same to Fast R-CNN. (Opinion) This data has multi-label structure. Note that input domain of loss is an anchor box, not anchor boxes sharing center. This design relaxes the issue about class correlation between anchor boxes. i is Seoul National University Deep Learning March-June, 2018 32 / 57

R-CNN Training faster R-CNN RPN and fast R-CNN share feature extractor part. This shared structure reduces test-time, the origin of its name Faster R-CNN. Sharing structure is implemented by following sequence. Phase Feature Extractor Region Proposal 1. Train RPN 2. Train fast R-CNN 3. Tune RPN 4. Tune fast R-CNN Initialized from ImageNet model Initialized from ImageNet model Frozen from phase 2. Frozen from phase 2. - RPN from phase 1. - RPN from phase 3. Seoul National University Deep Learning March-June, 2018 33 / 57

R-CNN Summary of R-CNN variations All the models use bbox regressor to adjust proposed region. R-CNN uses SVM and others use softmax classifier. Region Proposal Method Region Scaling Method R-CNN SS Warping Fast R-CNN SS RoI pooling Faster R-CNN RPN RoI pooling Table: Key methodologies. map (%) test time (ms/image) R-CNN 66.0 > 10 4 Fast R-CNN 66.9 1830 Faster R-CNN 69.9 198 Table: Evaluation on VOC 2007 test set, adjusted from Girshick and Ross, 2015. and Ren et al., 2015. Seoul National University Deep Learning March-June, 2018 34 / 57

YOLO YOLO Seoul National University Deep Learning March-June, 2018 35 / 57

YOLO You-Only-Look-Once You-Only-Look-Once (YOLO; Redmon et al., 2016.) is an unified approach model. YOLO has one CNNs solving both location and classification problem. In the introduction of paper: Humans glance at an image and instantly know what objects are in the image, where they are, and how they interact. For given image, YOLO feedforwards only one time, remarkably reducing test time. All the figures, tables, and equations in this section are come from Redmon et al., 2016. Seoul National University Deep Learning March-June, 2018 36 / 57

YOLO Terminology An image is divided by S S grid cells. If the center of an object falls into a grid cell, that grid cell is responsible for detecting that object. Seoul National University Deep Learning March-June, 2018 37 / 57

YOLO Terminology (Conti.) Each grid cell predicts B bounding boxes and corresponding objectness confidence. Each bounding box is parametrized by (x, y, w, h), the same to R-CNN. The objectness confidence is Pr(Object) IOU truth pred. Seoul National University Deep Learning March-June, 2018 38 / 57

YOLO Terminology (Conti.) All bounding boxes sharing grid cell have the same conditional class probability, formally Pr(Class i Object). At test time, the class-specific confidence, Pr(Class i ) IoUpred truth is predicted by multiplying predicted conditional class confidence and objectness confidence. Seoul National University Deep Learning March-June, 2018 39 / 57

YOLO Terminology (Conti.) Seoul National University Deep Learning March-June, 2018 40 / 57

YOLO Architecture For given image, YOLO predicts (x, y, w, h) and objectness confidence for all bounding boxes and conditional class probability for all grid cells. Considering its spatial meaning, we can view the output as S S (B 5 + C) box. In VOC competition, S = 7, B = 2, and C = 20. Seoul National University Deep Learning March-June, 2018 41 / 57

YOLO Architecture (Conti.) Following figure describes the architecture of YOLO. For given image, convolution layers extract features and final FC layer predicts bounding box parameters, objectness confidence, and conditional class probability. Seoul National University Deep Learning March-June, 2018 42 / 57

YOLO Loss Following is the loss function of YOLO. The first two terms are about bbox regression. Next two terms are about objectness classification and the last term is about the class classification. Here, 1 i and 1 ij are indicators about responsibility of ith grid cell and its jth bounding box, respectively. Seoul National University Deep Learning March-June, 2018 43 / 57

YOLO Performance YOLO is the first deep-learning model in the context of real-time detection with the state-of-the-art accuracy. Real-Time : 30 frames per second or better. When the speed of car is 60km/h, car moves 0.55m between detections. Seoul National University Deep Learning March-June, 2018 44 / 57

YOLO Performance (Conti.) Compared with fast R-CNN, YOLO has high location error and low background error. Correct: correct class and IoU >.5, Loc: correct class,.1<iou<.5, Sim: class is similar, IoU>.1, Other: class is wrong, IoU>.1, Background: IoU<.1 for any object. Seoul National University Deep Learning March-June, 2018 45 / 57

Evaluation Evaluation Seoul National University Deep Learning March-June, 2018 46 / 57

Evaluation Non Maximum Suppression Some of predicted regions severely overlap. In object detection, multiple detection for single GT is penalized. Figure: from https://kr.mathworks.com/help/vision/ref/selectstrongestbbox.html Seoul National University Deep Learning March-June, 2018 47 / 57

Evaluation Figure: from https://kr.mathworks.com/help/vision/ref/selectstrongestbbox.html. Seoul National University Deep Learning March-June, 2018 48 / 57 Non Maximum Suppression (Conti.) Non Maximum Suppression (NMS) is a pre-work for evaluation, removing overlapped regions using confidence. Choose the most confident bounding box and remove all other boxes with high IoU with the box. Repeat until there is no more box. NMS is applied to both RPN and detection network.

Evaluation Evaluation measures In Infinitely Imbalance Structure, performance measures using TN may unsuitable. True Predicted N P Total N TN FP # of BG P FN TP # of GT Table: The confusion matrix of object detection. Detecting all objects as it is easy. Just classify all regions to all object. What would be the value of TN? If the model is reasonable, TN should be. Seoul National University Deep Learning March-June, 2018 49 / 57

Evaluation Evaluation measures (Conti.) Main evaluation measures in object detection are based on Precision and Recall. Precision : the proportion of TP among positive labeled, TP/(TP+FP). Recall : the proportion of TP among positive, TP/(TP+FN). F1 score : the harmonic mean of precision and recall. AUCPR : the area under the precision-recall curve. Commonly used estimator for AUCPR in object detection is Average Precision (AP). Seoul National University Deep Learning March-June, 2018 50 / 57

Evaluation Average Precision Let c be the threshold of confidence. Than AUCPR can be expressed as AUCPR = Precision(c)dRecall(c) where Precision(c) and Recall(c) are the precision and recall at threshold level c, respectively. By plugging in the empirical precision and recall, Precision(c) and Recall(c), we get an estimator of AUCPR, AUCPR = Precision(c)d Recall(c). Seoul National University Deep Learning March-June, 2018 51 / 57

Evaluation Average Precision (Conti.) Here, by the definition of Riemann Stieltjes integral, AUCPR = = Precision(c)d Recall(c) c {conf i i P} Precision(c) ( # of P have conf. equal to c ). # of P This is the Average Precision (AP), an weighted average of precisions at each confidence level of GT box. Seoul National University Deep Learning March-June, 2018 52 / 57

Evaluation Average Precision (Conti.) Considering various kinds of class, the mean of AP is used. This is called mean Average Precision (map). In practice, Interpolated AP is used due to the wiggles in the precision-recall curve. Unlike the ROC curve, it may not hold monotonicity. Seoul National University Deep Learning March-June, 2018 53 / 57

Evaluation References Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 580-587). Uijlings, Jasper RR, et al. Selective search for object recognition. International journal of computer vision 104.2 (2013): 154-171. Felzenszwalb, Pedro F., and Daniel P. Huttenlocher. Efficient graph-based image segmentation. International journal of computer vision 59.2 (2004): 167-181 Felzenszwalb, Pedro F., et al. Object detection with discriminatively trained part-based models. IEEE transactions on pattern analysis and machine intelligence 32.9 (2010): 1627-1645. Seoul National University Deep Learning March-June, 2018 54 / 57

Evaluation References (Conti.) A. Krizhevsky, I. Sutskever, and G. Hinton. ImageNet classification with deep convolutional neural networks. In NIPS, 2012. He, Kaiming, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition. european conference on computer vision. Springer, Cham, 2014. Girshick, Ross. Fast r-cnn. arxiv preprint arxiv:1504.08083 (2015). Simonyan, Karen, and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arxiv preprint arxiv:1409.1556 (2014). Seoul National University Deep Learning March-June, 2018 55 / 57

Evaluation References (Conti.) Ren, Shaoqing, et al. Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems. 2015. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788). Everingham, M., Van Gool, L., Williams, C. K., Winn, J., and Zisserman, A. (2010). The pascal visual object classes (voc) challenge. International journal of computer vision, 88(2), 303-338. Seoul National University Deep Learning March-June, 2018 56 / 57

Evaluation References (Conti.) Boyd, Kendrick, Kevin H. Eng, and C. David Page. Area under the precision-recall curve: Point estimates and confidence intervals. Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, Berlin, Heidelberg, 2013. Introduction to modern information retrieval Seoul National University Deep Learning March-June, 2018 57 / 57