Главная
Study mode:
on
1
Intro
2
Self-Driving - A Human Dream
3
Imitation Learning
4
Sensors
5
Geometric Fusion Lacks Global Context
6
TransFuser: Key Idea
7
TransFuser. Full Architecture
8
TransFuser: Loss Functions
9
TransFuser. Experimental Evaluation
10
TransFuser: Results on Longest6 Benchmark
11
TransFuser. Results on CARLA Leaderboard
12
TransFuser. Attention Maps
13
TransFuser. Failures
14
Rule-based Planner
15
End-to-End Driving Policy
16
Learned Planner
17
PlanT: Intermediate Representation
18
PlanT: Model
19
PlanT: Input Representation
20
KING: Critical Scenarios as Attacks
21
KING: Overview
22
KING: Gradient Paths
23
KING: Cost Functions
24
KING: Collision Types
25
KING: Adversarial Cut-in Maneuver
26
Summary
Description:
Explore a comprehensive talk from the ECCV 2022 workshop on Autonomous Vehicle Vision, focusing on robust policies for self-driving vehicles. Delve into three key projects: TransFuser, PlanT, and KING. Examine the challenges of geometric fusion, the benefits of global context in perception, and innovative approaches to imitation learning. Investigate the architecture, loss functions, and experimental results of TransFuser, including its performance on benchmarks and the CARLA Leaderboard. Learn about end-to-end driving policies, intermediate representations in learned planners, and the concept of critical scenarios as attacks in autonomous driving. Gain insights into gradient paths, cost functions, and collision types in adversarial scenarios, concluding with a summary of cutting-edge advancements in self-driving technology.

Learning Robust Policies for Self-Driving

Andreas Geiger
Add to list
0:00 / 0:00