Completed
Panoptic-DeepLab
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Segmenting and Tracking Every Pixel
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Overview
- 3 Why Segmentation matters
- 4 Why Tracking matters
- 5 Evolution of Visual Scene Understanding
- 6 From Tracking to STEP
- 7 Contributions
- 8 Existing Datasets
- 9 Annotation Process
- 10 Dataset Comparison (Training Set)
- 11 Tracking the most salient classes
- 12 Track length distribution
- 13 Why a new metric?
- 14 Metric design
- 15 Formal Definition
- 16 Association Quality (AQ)
- 17 Segmentation Quality (SQ) We use Intersection-over-Union (loU) to measure segmentation quality
- 18 Segmentation and Tracking Quality (STO)
- 19 Panoptic-DeepLab
- 20 Extensions to Tracking Single-frame baselines
- 21 Results on KITTI-STEP
- 22 Resources
- 23 Summary
