PPT-Summary of “Efficient Deep Learning for Stereo Matching”

Xi Mo 432017 x y z x y xl xr Disparityxl xr Disparity space Traditional way of stereo matching Benchmark of Middlebury 3DMSTRank 1 Error Map Disparity map of left view Left view. Many slides adapted from Steve Seitz. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. image 1. image 2. Dense depth map. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. Packet Inspection of Next Generation Network Devices. . Prof. Anat Bremler-Barr. IDC . Herzliya. www.deepness-lab.org. This work was supported by European Research Council (ERC) Starting Grant no. 259085 , . Chapter 11 Stereo Correspondence. Presented by: . 蘇唯誠. 0921679513. r02922114@ntu.edu.tw. 指導教授. : . 傅楸善 博士. Introduction. Stereo matching is the process of taking two or more images and estimating a 3D model of the scene by finding matching pixels in the images and converting their 2D positions into 3D depths.. Many slides adapted from Steve Seitz. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. image 1. image 2. Dense depth map. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. Many slides adapted from Steve Seitz. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. Where does the depth information come from?. Binocular stereo. Given a calibrated binocular stereo pair, fuse it to produce a depth image. Many slides drawn from Lana . Lazebnik. , UIUC. Basic stereo matching algorithm. For each pixel in the first image. Find corresponding . epipolar. line in the right image. Examine all pixels on the . Noah . Snavely. , . Zhengqi. Li. Single image stereogram, by . Niklas. . Een. Mark Twain at Pool Table", no date, UCR Museum of Photography. Stereo. Given two images from different viewpoints. How can we compute the depth of each point in the image?. optimisation. Milica. Ga. š. i. ć. Dialogue Systems Group. Structure of spoken . dialogue systems. Language understanding. Language generation. semantics. a. ctions. 2. Speech recognition. Dialogue management. Network to Compare Image Patches. Jure . Zbontar. , Yann . LeCun. Background. Motivation. Problem Formulation. Methodology. Training Data. Suggested Net Architectures. Sequential Steps. Results. Conclusion. Kexin Pei. 1. , Yinzhi Cao. 2. , Junfeng Yang. 1. , Suman Jana. 1. 1. Columbia University, . 2. Lehigh University. 1. Deep learning (DL) has matched human performance!. Image recognition, speech recognition, machine translation, intrusion detection.... 6.819 / 6.869: Advances in Computer Vision. Antonio Torralba. Camera Models. ?. ?. Perspective projection. Virtual image plane. (0,0,0). Perspective projection. (0,0,0). f. Z. Y. y. ?. y. = . f. Y/Z. Several slides from Larry . Zitnick. and Steve Seitz. Why do we perceive depth?. What do humans use as depth cues?. Convergence . When watching an object close to us, our eyes point slightly inward. This difference in the direction of the eyes is called convergence. This depth cue is effective only on short distances (less than 10 meters). . 1. 2. Why do we perceive depth?. 3. What do humans use as depth cues?. Convergence . When watching an object close to us, our eyes point slightly inward. This difference in the direction of the eyes is called convergence. This depth cue is effective only on short distances (less than 10 meters). . Several slides from Larry . Zitnick. and Steve Seitz. Why do we perceive depth?. What do humans use as depth cues?. Convergence . When watching an object close to us, our eyes point slightly inward. This difference in the direction of the eyes is called convergence. This depth cue is effective only on short distances (less than 10 meters). .