CS5720 - Week 10
Slide 181 of 200

Computer Vision Problem Types

🖼️

Image Classification

Assigning a label to an entire image from a predefined set of categories

Cat vs Dog CIFAR-10 ImageNet
📦

Object Detection

Locating and classifying multiple objects within an image using bounding boxes

YOLO R-CNN SSD
🎨

Image Segmentation

Classifying each pixel in an image to identify object boundaries precisely

Semantic Instance Panoptic
👤

Face Recognition

Identifying or verifying individuals based on facial features

FaceNet DeepFace ArcFace
🎭

Style Transfer

Applying artistic styles from one image to the content of another

Neural Style Fast Style CycleGAN
🔍

Super-Resolution

Enhancing image quality by increasing resolution and adding details

SRCNN ESRGAN Real-ESRGAN

Complexity & Computational Requirements

Low
Classification
Med
Detection
High
Segmentation
Prepared by Dr. Gorkem Kar