BlindSpotNet: Seeing Where We Cannot See

TL;DR

BlindSpotNet is a novel approach that estimates road blind spots from monocular images to enhance road scene understanding and safety, using a large-scale automatically generated dataset and a dedicated neural network.

Contribution

We propose a new method for 2D blind spot estimation from monocular cameras, including automatic dataset creation and a neural network model, addressing challenges of 3D reasoning and real-time detection.

Findings

The RBS dataset effectively trains blind spot estimation models.

BlindSpotNet accurately predicts blind spots in diverse driving scenarios.

Our approach outperforms existing methods in blind spot detection accuracy.

Abstract

We introduce 2D blind spot estimation as a critical visual task for road scene understanding. By automatically detecting road regions that are occluded from the vehicle's vantage point, we can proactively alert a manual driver or a self-driving system to potential causes of accidents (e.g., draw attention to a road region from which a child may spring out). Detecting blind spots in full 3D would be challenging, as 3D reasoning on the fly even if the car is equipped with LiDAR would be prohibitively expensive and error prone. We instead propose to learn to estimate blind spots in 2D, just from a monocular camera. We achieve this in two steps. We first introduce an automatic method for generating ``ground-truth'' blind spot training data for arbitrary driving videos by leveraging monocular depth estimation, semantic segmentation, and SLAM. The key idea is to reason in 3D but from 2D images by defining blind spots as those road regions that are currently invisible but become visible in the near future. We construct a large-scale dataset with this automatic offline blind spot estimation, which we refer to as Road Blind Spot (RBS) dataset. Next, we introduce BlindSpotNet (BSN), a simple network that fully leverages this dataset for fully automatic estimation of frame-wise blind spot probability maps for arbitrary driving videos. Extensive experimental results demonstrate the validity of our RBS Dataset and the effectiveness of our BSN.