Object Detection

Object detection.

• Region Proposal Networks: R-CNN, Fast R-CNN, Faster R-CNN.
• Single Shot Detectors: SSD, RetinaNet, YOLO.

Materiale didattico:

• slides
• Notebook di accompagnamento
• Demo YOLO

Riferimenti bibliografici

• [Elg20], Ch. 7

Link utili

• An overview of Deep-Learning based Object Detection algorithms

• Object detection with Deep Learning: A review

• Rich Feature-Based hierarchies for accurate object detection and semantic segmentation

• Fast R-CNN

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

• Faster R-CNN tutorials: A guide to building Faster R-CNN in Pytorch e Detecting objects in (almost) Real time

• SSD: Single Shot MultiBox Detector

• SSD Tutorial

• Detectron and Detectron2

• You Only Look Once: YOLOv1, YOLOv2, YOLOv3

• Understanding YOLO

• What’s new in YOLOv3?

• YOLOv3 from Scratch

• Speed and accuracy comparison in object detection

• YOLOv4 paper (with code)

• YOLOv4 - Superior, faster and more accurate

• YOLOv4 in Pytorch

• YOLOv5: The Latest Model for Object Detection

• Focal Loss for Dense Object Detection

• Review - RetinaNet

• RetinaNet explained and demystified

• 5 tools to create a custom object detection dataset

• NEW DETR: End-to-end object detection with Transformers

• NEW EfficientDet