Buy Me a Coffee☕

*My post explains Image Classification(Recognition), Object Localization, Object Detection and Image Segmentation.

Computer Vision is the technology which enables a computer to understand and analyze the visual things such as images, videos, etc.

(1) Keypoint Detection(Landmark Detection):

• can localize the distinctive points of the objects or interest regions in an image: *Memos:
  • The image can be one frame in a video.
  • Object Localization or Image Classification and Object Localization(Object Detection) is used.
• can is used to recognize face, its emotion, the pose of person, clothes, etc.

(2) Image Matching:

• can determine that the objects and interest regions in multiple images are the same or similar by comparing their distinctive points.
• is the advanced technology of Keypoint Detection.
• can be used to match faces, finger prints, etc.

(3) Object Tracking:

• can track the objects and interest regions in a video(multiple frames captured in sequence) with bounding boxes. *The video can be multiple images captured in sequence.
• is the advanced technology of Object Detection.
• is used for Surveillance, Autonomous Vehicles, Virtual Reality, Manufacturing, Unmanned Stores, etc.

(4) Optical Flow:

• can detect and visualize the motion(velocity) of the objects and interest regions in a video(multiple frames captured in sequence): *Memos:
  • Velocity is the spped and direction of a moving object.
  • The video can be multiple images captured in sequence.
• is the branched technology of Object Tracking.
• is used for Surveillance, Traffic Monitoring, Motion Analysis, etc.

(5) Stereo Matching:

• can create a 3D scene from two or more 2D stereo images by computing disparity: *Memos:
  • A 2D stereo image is the similar image taken from a different position.
  • Disparity is the distance between two or more corresponding pixels in the stereo images.
• is used for Robotics.

(6) Video Prediction can predict future video frames from past video frames.

(7) Image Captioning:

• can generate textual descriptions for images.
• is the generative AI which is the combination of Computer Vision and Natural Language Processing(NLP).