Give a counter-example in which apparent motion measured by...

Name: Problem 2, Chapter 12: Image processing, Analysis, and Machine Vision
Uploaded: 2021-12-12T05:03:17-08:00
Duration: 2 min 30 s
Channel: Manish Jain
Description: Give a counter-example in which apparent motion measured by an optical flow fiek does not correspond to actual motion of the 3D points. [Hint: Consider an opaque sphere, where (1) the sphere does not move and illumination changes; (2) the sphere rotates and the illumination is constant.]

Give a counter-example in which apparent motion measured by an optical flow fiek does not correspond to actual motion of the 3D points. [Hint: Consider an opaque sphere, where (1) the sphere does not move and illumination changes; (2) the sphere rotates and the illumination is constant.]

Key Concepts

Surface Rotation and Shading Effects

When a three-dimensional object rotates, even with constant illumination, the way light interacts with its surface can change due to varying angles of incidence and reflective properties. This results in changes in shading and highlights over time. Optical flow algorithms can misinterpret these changes as motion, even though the object’s physical position in space may not correspond to what the apparent motion field indicates.

Illumination Changes

Illumination changes refer to variations in lighting conditions that can alter the appearance of an object without any actual motion occurring. Such changes can falsely suggest movement in the computed optical flow field because the intensity variations produced by lighting fluctuations are misinterpreted as motion, even when the object itself is stationary.

Brightness Constancy Assumption

The brightness constancy assumption is a fundamental concept in optical flow that holds that the brightness of a point in the image remains constant between successive frames. This assumption simplifies the calculation of motion, but it can be violated when lighting conditions change or when the object rotates in a way that alters the shading, leading to discrepancies between the measured optical flow and the true 3D motion.

Optical Flow

Optical flow is a technique used in computer vision to estimate the apparent motion of objects between successive image frames by tracking changes in brightness patterns. It is based on the movement of pixel intensities in the image, and while it often corresponds to the actual motion of objects, it can be affected by factors such as illumination changes or changes in object appearance due to rotation.

Transcript

Please give Ace some feedback