Author: KnowingAI knowledge and wisdom
link: https: //www.zhihu.com/question/323644172/answer/715321371
Source: know almost
copyrighted by the author. Commercial reprint please contact the author authorized, non-commercial reprint please indicate the source.
When it comes to computer vision, necessarily leading Daniel David Marr (1945-1980).
In the 1970s, as scientists Marr nervous system, the knowledge neuropsychology and artificial intelligence proposed by combining the cross-era significance of "visual computing theory", marking the computer vision to become an independent discipline.
In Marr view, the main function of the visual is "from two-dimensional images of the retina imaging, three-dimensional object surface shape of the recovery," which is "reconstruction." He also made a series of theories and methods of calculation about the "reconstruction" of.
To commemorate the pioneering contributions to the field of computer vision Marr made, the highest honor in computer vision "Mull Award" to his name. Who once wrote the ICCV (International Conference of Computer Vision, The International Conference on Computer Vision), "Marr Award" will be awarded the most outstanding paper in two years.
Marr works quite good, but it's limited, many now suspect Marr have been proved incorrect. His theory is summarized by its students a "vision", are interested can get a look.
Looking for "big cow" can start from the top conferences in the field of computer vision.
Chairing meetings at these top judges, or papers to be adopted by scholars, it is worth learning "seniors."
In addition to ICCV previously mentioned, there CVPR (Computer Vision and Pattern Recognition Conference, IEEE Computer Vision and Pattern Recognition), ECCV (European Conference on Computer Vision, European Conference on Computer Vision), and so on.
(Top image will be on the field, there is an answer for reference: picture field which advanced with the top journals will? )
Such as:
1.Takeo Kanade (Takeo Kanade)
Professor of Computer Science and Robotics Institute at Carnegie Mellon University, dean of computer vision, the US Academy of Engineering. The research focused on the target tracking, face detection and scene analysis, 3D, KLT tracking algorithm in the "K" refers to him. Kanade has won awards in Mar 1990, In addition, he also holds dozens of patents.
2.Andrew Zisserman
Oxford University professor and director of vision measurement group. Had in 1993, 1998 and 2003, three-time Markov prize. Author of computer vision classic book "Multiview Visual Geometry", also a famous deep learning network: One of the authors VGGNet (Visual Geometry Group Net) is.
3.David Lowe
Professor, Department of Computer Science, University of British Columbia, in 1988 the study of human visual computing model and won the prize Malta, focusing on target identification. 1999, an image processing algorithm known SIFT Algorithm (scale-invariant feature transform) David invention, became the most popular method for image feature detection, object recognition, and is widely used in areas such as image classification.
4.Paul Viola
The most influential in terms of human face detection scientist, in 2003 by the Maltese award. In the 2001 CVPR, he and Micheal Jones co-published paper "Rapid object detection using a boosted cascade of simple features", was influenced scholars a great extent, so boosting cascade algorithm which has been widely used in academic and industry.
5.Andrew Blake
2 ECCV won the Best Paper Award and one Award for Malta. Target tracking and tracing aspects of body posture focused on image segmentation. The famous Active Shape Model (Active Shape Model) is by his participation invention algorithm "Active Contours' evolved.
6.Luc Van Gool
Belgium Dutch professor at the Catholic University of Leuven, Swiss Federal Institute of Technology Zurich, Professor, Computer Vision Laboratory, 1998 by Mar award. On the basis of his SIFT algorithm, we developed a more advanced SURF algorithm (speed up robust feature).
7.David Forsyth
American University of Illinois, Urbana - Champaign, Professor, Chairman of the 2019 ICCV General Assembly in 1993 by Marr Prize, and Jean Ponce co-author of "Computer Vision: A Modern Approach" classic textbook called computer vision.
8. Zhu Songchun
Computer Vision famous Chinese scholar, now a professor at UCLA Department of Statistics and Computer Science, University of California, UCLA computer vision, cognition, learning and director of the Center for autonomous robots 2019 CVPR President. Zhu Songchun won 1st prize Malta, understanding the scene, the visual aspects of common sense reasoning made important contributions.
9.Marc Pollefeys
ETH Zurich professor, President of the General Assembly in 2019 ICCV, self-calibration of the camera has made an outstanding contribution. He published with two other scholars on a IJCV (International Journal of Computer Vision, International Journal of Computer Vision) article "Self-Calibrationand metric reconstruction inspite of varying and unknown intrinsic camera parameters", is called self-calibration aspects most one of the classic papers, and thus he won Marr Prize in 1998.
10.Larry Davis
University of Maryland professor, President of the General Assembly in 2019 CVPR. Its most famous research results W4 real-time monitoring system - the first real-time video surveillance system capable of detecting and outdoor human behavior analysis.
11.Phillip Torr
British professor at Oxford University, President of the General Assembly CVPR 2019, had won the 1998 prize Mar, now the autopilot FiveAI company's chief scientific adviser.
12.Kyoung Mu Lee
Seoul National University professor of international CV very active community of Korean scholars, 2019 ICCV the President of the General Assembly, the field of prestigious journals TPAMI associate editor.
13.Gerard Medioni
President of the General Assembly in 2020 CVPR, who co-authored "Emerging Topics of Computer Vision", "multimedia systems: algorithms, and standard industry practice" and other classic academic books.
Contact some "big cow" and not so top-level meeting "close", but along the development context and the position of computer vision, you can still find them.
Such as:
1.Jitendra Malik
Stanford University professor whose research work has focused on contour detection, image / video segmentation, pattern matching, etc., have outstanding contributions in edge detection, image segmentation and shape matching terms. Malik talents of students, Chinese scholars Shi Jianbo to come from his door, also includes Alexie Efros, Paul Debevec, Yair Weiss and other outstanding scholars.
2.David Murray
British professor at Oxford University, Active Vision Laboratory Director, inventor PTAM algorithm (Parallel tracking and mapping) of. Philip Torr He is mentioned above, Andrew Davison's PhD supervisor.
3.Andrew Davison
Imperial College London Professor Dyson machine vision and robotics laboratory director of the group, focusing on research in the field of SLAM, for the advent of MonoSLAM, SLAM ++ and other achievements have made a great contribution.
4.Vijay Kumar
University of Pennsylvania professor, Vijay Kumar, director of the laboratory, have high achievements in the field of unmanned aerial vehicles.
5.Daniel Cremers
Professor Technical University of Munich, Germany, Computer Vision Group Director. Also outstanding scholars in the field of UAVs, or one of the inventors LSD-SLAM algorithm.
6.Richard Szeliski
His book "Computer Vision: Algorithms and Applications" (Computer Vision: Algorithms and Applications) computer vision is also very classic introductory textbook.
Add some well-known Chinese scholars -
1. Thomas Huang
University of Illinois at Urbana - Champaign (UIUC) Beckman Research Institute of the image lab director, US National Academy of Engineering. Focus on research aspects of image processing, pattern recognition, computer vision and human-computer interaction, and the invention of the predicted differential quantization, trained many outstanding students.
2. Shi Jianbo
Computer and Information Science, University of Pennsylvania professor, a high achievement in terms of image segmentation. In 2000, Shi Jianbo and others in the journal article PAMI jointly issued "Normalized cuts and image segmentation" breakthrough graph theory method is applied to image segmentation and clustering, making it a mainstream research methods to promote a great extent, image segmentation, as well as the development of the field of computer vision.
3. Li Feifei
Stanford University professor, director of Artificial Intelligence Laboratory and visual laboratory. ImageNet founder of the famous race. From 2010 to 2017, the annual competition has attracted many outstanding scholars, also witnessed the rapid development of image recognition algorithm in 2017, winning recognition accuracy of the algorithm even reached 97.3% higher than the human level.
Li Feifei also won the 2018 "World Chinese influence Award", also has a considerable reputation among less concerned about the AI field.
4. Tangxiao Ou
Chinese University of Hong Kong Department of Information Engineering, with a high achievement in face recognition.
In 2009, the paper won the team Tangxiao Ou annual "Best Paper Award" at the CVPR, is CVPR since its inception 25 years for the first time since winning the Asian papers. IJCV first Chinese editor; in 2014, GaussianFace algorithm published by Tangxiao Ou team, the accuracy rate in the world's most authoritative face recognition database LFW of 98.52%, for the first time beyond the human level.
Of course, he has another identity: Technology Shang Dynasty founder.
5. Wu Ying
Professor at Northwestern University, Computer Vision Laboratory. Professor of American youth won the highest honor: the US National Science Foundation "Career Award", and has made outstanding achievements in gesture recognition and target tracking.
6. Sun Jian
Masterpieces of Science and Technology's chief scientist, research interests include face recognition, computational photography and image understanding based on the depth of learning. Since 2002, he published more than 100 papers in SIGGRAPH, CVPR and other top-level meeting, the two-time CVPR Best Paper Award. His team developed the depth of residual network (ResNet) has a broad impact on the field of image recognition.
7. Jia Jaya
Chinese University of Hong Kong, Tencent excellent map laboratory scientist. The main research interests include computer vision, image processing, pattern recognition, made a great contribution in terms of image filtering, image processing sparse, multi-band image signal fusion.
Hope that helps -
