AutoLay: Benchmarking amodal layout estimation for autonomous driving

TL;DR

AutoLay introduces a comprehensive benchmark dataset for amodal layout estimation in autonomous driving, combining data from KITTI and Argoverse with standardized evaluation protocols.

Contribution

It provides a standardized dataset and benchmark for amodal layout estimation, addressing inconsistencies in task definitions and evaluation methods.

Findings

Baseline and advanced methods evaluated on AutoLay.

AutoLay includes detailed semantic annotations and 3D point clouds.

The benchmark facilitates fair comparison of amodal layout estimation approaches.

Abstract

Given an image or a video captured from a monocular camera, amodal layout estimation is the task of predicting semantics and occupancy in bird's eye view. The term amodal implies we also reason about entities in the scene that are occluded or truncated in image space. While several recent efforts have tackled this problem, there is a lack of standardization in task specification, datasets, and evaluation protocols. We address these gaps with AutoLay, a dataset and benchmark for amodal layout estimation from monocular images. AutoLay encompasses driving imagery from two popular datasets: KITTI and Argoverse. In addition to fine-grained attributes such as lanes, sidewalks, and vehicles, we also provide semantically annotated 3D point clouds. We implement several baselines and bleeding edge approaches, and release our data and code.