Completed
State-of-the-Art Vision-based Object Detection
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Perception Deception - Physical Adversarial Attack Challenges
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Car Safety - Unintended Acceleration
- 3 Car Perception While Driving
- 4 Behind Perception: End2End Object Detection
- 5 State-of-the-Art Vision-based Object Detection
- 6 Current Status of Adversarial Example
- 7 Adversarial Examples & L2 Norm Perturbations Impact to DNN
- 8 Explore Chances of Physical White Box Attack against YOLOV3
- 9 Deep Dive into YOLOv3
- 10 Training Dataset - MS COCO Dataset
- 11 YOLOv3 Prediction
- 12 Threat Model : Physical Image Patch Attack
- 13 Our Physical Attack Approach & Objectives
- 14 Differentiable Input Patch Construction
- 15 Attack Objective 1 - Object Fabrication
- 16 Attack Objective 2 - Object Vanishing
- 17 Challenges to the Success of Physical Attack
- 18 Tactics to the Challenges
- 19 Color Management with Non-Printability Loss
- 20 RT + TV for Various Distances & Angles
- 21 Put Everything Together: An Iterative Optimization
- 22 Conclusion & Takeaway
