AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Think Twice Before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving

Xiaosong Jia, Penghao Wu, Li Chen, Jiangwei Xie, Conghui He, Junchi Yan, Hongyang Li

CVPR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 111 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use supervised depth and segmentation predictions during image-to-BEV projection (ablation)ThinkTwice encoder · Town05-LongDS: 43.0$\pm$3.9 -> 57.8$\pm$1.5Improved
Add image depth and segmentation auxiliary tasks (ablation)ThinkTwice encoder · Town05-LongDS: 43.0$\pm$3.9 -> 45.0$\pm$2.2Improved
Use two input frames instead of one (ablation)ThinkTwice encoder · Town05-LongDS: 57.8$\pm$1.5 -> 59.3$\pm$1.4Improved
Use expert BEV feature distillation instead of BEV segmentation supervision (ablation)ThinkTwice encoder · Town05-LongDS: 50.5$\pm$2.3 -> 59.3$\pm$1.4Improved
Add one proposed decoder module (ablation)ThinkTwice · Town05-LongDS: 59.3$\pm$1.4 -> 61.6$\pm$1.4Improved
Stack five proposed decoder modules (ablation)ThinkTwice · Town05-LongDS: 59.3$\pm$1.4 -> 65.0$\pm$1.7Improved
Add the Look Module (ablation)ThinkTwice decoder · Town05-LongDS: 62.4$\pm$1.8 -> 65.0$\pm$1.7Improved
Add the Prediction Module (ablation)ThinkTwice decoder · Town05-LongDS: 61.7$\pm$1.9 -> 65.0$\pm$1.7Improved
Train the Prediction Module with teacher forcing (ablation)ThinkTwice decoder · Town05-LongDS: 62.0$\pm$2.2 -> 65.0$\pm$1.7Improved
Increase the TCP encoder from ResNet-34 to ResNet-101 (ablation)TCP · Town05-LongDS: 57.2$\pm$1.5 -> 38.0$\pm$4.5Hurt
Increase TCP encoder from ResNet-34 to ResNet-18 (ablation)TCP · Town05-LongDS: 57.2$\pm$1.5 -> 48.7$\pm$1.7Hurt
Increase the TCP MLP/GRU decoder depth and width (ablation)ThinkTwice baseline · Town05-LongDS: 59.3$\pm$1.4 -> 59.4$\pm$2.4No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy