PPT-Rich feature Hierarchies for Accurate object detection and

Ross Girshick Jeff Donahue Trevor Darrell Jitandra Malik UC Berkeley Presenter Hossein Azizpour Abstract Can CNN improve soa object detection results Yes it helps by learning rich representations which can then be combined with computer vision techniques. Video Analytics. Why Video Analytics?. The increasing rate of crime calls for effective security measures.. Security Personnel, IP Cameras, CCTV are usually employed for these reasons.. But Human vigilance is required in each case which is bound to induce errors. . Collisions. Collisions. Detection. Broad Phase. Bounding Volumes. Key idea:. Surround the object with a (simpler) bounding object (the bounding volume).. If something does not collide with the bounding volume, it does not collide with the object inside.. Image Processing . Pier Luigi . Mazzeo. pierluigi.mazzeo@. cnr.it. Find. Image . Rotation. and Scale Using . Automated. . Feature. . Matching. and RANSAC. Step. 1: Read . Image. original. = . Binarized. Normed Gradients for . Objectness. Estimation at 300fps. CVPR 2014 Oral. Outline. 1. . Introduction. 2.. . Methodology. 2.1 Normed . gradients (NG) and . objectness. 2.2 Learning . objectness. Oscar . Danielsson. (osda02@kth.se). Stefan . Carlsson. (. stefanc@kth.se. ). Outline. Detect all Instances of an Object Class. The classifier needs to be fast (on average). This is typically accomplished by:. for Object Detection. Forrest Iandola, . Ning. Zhang, Ross . Girshick. , Trevor Darrell, and Kurt . Keutzer. Deformable Parts Model (DPM): state of the art algorithm for object detection [1]. Several attempts to accelerate multi-category DPM detection, such as [2] [3]. Image Processing. Pier Luigi Mazzeo. pierluigi.mazzeo@cnr.it. Image Rotation &. Object . Detection . Find. Image . Rotation. and Scale Using . Automated. . Feature. . Matching. and RANSAC. Step. Before deep . convnets. Using deep . convnets. PASCAL VOC. Beyond sliding windows: Region proposals. Advantages:. Cuts . down on number of regions detector must . evaluate. Allows detector to use more powerful features and classifiers. Yao Lu, Linda Shapiro. University of Washington. AAAI-17. Background: human visual perception. Object perception. Edge perception. Assigning edges to regions. Grouping regions to objects. Bottom-up and top-down pathways. Facebook AI Research. Wenchi. Ma. Data: 11/04/2016. More information from object detection. More information from object detection. More information from object detection. Object Detection for now with Deep Learning. Oscar . Danielsson. (osda02@csc.kth.se). Stefan . Carlsson. (. stefanc@csc.kth.se. ). Josephine Sullivan (. sullivan@csc.kth.se. ). DICTA08. The Problem. Object categories are often modeled by collections (bag-of-features) or constellations (pictorial structures) of local features . Before deep . convnets. Using deep . convnets. PASCAL VOC. Beyond sliding windows: Region proposals. Advantages:. Cuts . down on number of regions detector must . evaluate. Allows detector to use more powerful features and classifiers. HOGgles Visualizing Object Detection Features C. Vondrick , A. Khosla , T. Malisiewicz , A. Torralba ICCV , 2013 . presented by Ezgi Mercan Object Detection Failures Why do our detectors think water looks like a car? Ming-Ming Cheng. 1. Ziming Zhang. 2. Wen-Yan Li. 1. Philip H. S. Torr. 1. 1. Torr . Vision Group, Oxford . University . 2. Boston . University. 1. Motivation: Generic . object detection.