Is Computer Vision Possible?
نویسنده
چکیده
| This paper summarises the rst session of the Vision Milestones'95 workshop. Key questions concerning the topic of the session are formulated. Personal views of the author|the session chairman| on some general problems facing computer vision are presented in a rather informal way. The main issues addressed are the ultimate goal of computer vision, computer versus human vision, heuristics versus theory, and signiicance and veriication of experiments. Then the author's brief answers to the key questions are given. Finally, the opinions of other participants are presented. 1 Key Questions Q1: (ultimate goal) What is the ultimate goal of Computer Vision (CV)? Is it replication of human vision capabilities, creating theories that support development of CV algorithms, or something else? Q2: (current goals) What are the current goals of computer vision? How do they relate to the ultimate goal? Q3: (computational structures) Are the existing computational structures adequate to vision? Q4: (experience and intuition) In human vision, what is the role of experience and intuition? At what stages do they enter the process of vision? Can we use it in computer vision? Q5: (reliability) Why are CV algorithms/systems so unreliable? Can they be robust? What degree of control over environment do we need for this? Q6: (ill-posedness) Many vision problems are ill-posed. How does human vision cope with this? Can we do it in computer vision? 1.1 Introduction The theory of computer vision is usually expected to provide principles by which algorithms can be designed to solve various automation tasks related to processing of visual information 1]. If computer vision is to become a \hard science", these principles should be derived from the initial problem statement, assumptions about the particular vision phenomenon considered, and the general laws of this science. A theory of computer vision must have predictive power, i.e. be able to predict the behavior of an algorithm in given circumstances.
منابع مشابه
Computer assisted instruction during quarantine and computer vision syndrome
Computer vision syndrome (CVS) is a set of visual, ocular, and musculoskeletal symptoms that result from long-term computer use. These symptoms include eyestrain, dry eyes, burning, pain, redness, blurred vision, etc, which increase with the duration of computer use. Currently, with the closure of schools and universities due to the continued COVID19 pandemic many universities have taken the pr...
متن کاملHuman Computer Interaction Using Vision-Based Hand Gesture Recognition
With the rapid emergence of 3D applications and virtual environments in computer systems; the need for a new type of interaction device arises. This is because the traditional devices such as mouse, keyboard, and joystick become inefficient and cumbersome within these virtual environments. In other words, evolution of user interfaces shapes the change in the Human-Computer Interaction (HCI). In...
متن کاملHuman Computer Interaction Using Vision-Based Hand Gesture Recognition
With the rapid emergence of 3D applications and virtual environments in computer systems; the need for a new type of interaction device arises. This is because the traditional devices such as mouse, keyboard, and joystick become inefficient and cumbersome within these virtual environments. In other words, evolution of user interfaces shapes the change in the Human-Computer Interaction (HCI). In...
متن کاملAn Investigation of Structural-Mechanical Properties of Spun-Bonded Non-Woven Using Computer Vision Method
متن کامل
Quad-pixel edge detection using neural network
One of the most fundamental features of digital image and the basic steps in image processing, analysis, pattern recognition and computer vision is the edge of an image where the preciseness and reliability of its results will affect directly on the comprehension machine system made objective world. Several edge detectors have been developed in the past decades, although no single edge detector...
متن کاملQuad-pixel edge detection using neural network
One of the most fundamental features of digital image and the basic steps in image processing, analysis, pattern recognition and computer vision is the edge of an image where the preciseness and reliability of its results will affect directly on the comprehension machine system made objective world. Several edge detectors have been developed in the past decades, although no single edge detector...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2007