Robust Environment Perception for Automated Driving: A Unified Learning Pipeline for Visual-Infrared Object Detection

IEEE Intelligent Vehicles Symposium, pp. 367–374

Abstract

The RGB complementary metal-oxide-semiconductor (CMOS) sensor works within the visible light spectrum. Therefore it is very sensitive to environmental light conditions. On the contrary, a long-wave infrared (LWIR) sensor operating in 8-14 μm spectral band, functions independent of visible light.In this paper, we exploit both visual and thermal perception units for robust object detection purposes. After delicate synchronization and (cross-) labeling of the FLIR [1] dataset, this multi-modal perception data passes through a convolutional neural network (CNN) to detect three critical objects on the road, namely pedestrians, bicycles, and cars. After evaluation of RGB and infrared (thermal and infrared are often used interchangeably) sensors separately, various network structures are compared to fuse the data at the feature level effectively. Our RGB-thermal (RGBT) fusion network, which takes advantage of a novel entropy-block attention module (EBAM), outperforms the state-of-the-art network [2] by 10% with 82.9% mAP.

Authors 5

1. Sam Vadidar
   Affiliation as printed

   Elektronische Fahrwerksysteme GmbH,Gaimersheim,Germany,85080

2. Ali Kariminezhad
   Affiliation as printed

   Elektronische Fahrwerksysteme GmbH,Gaimersheim,Germany,85080

3. Christian Mayr
   Affiliation as printed

   Elektronische Fahrwerksysteme GmbH,Gaimersheim,Germany,85080

4. Laurent Kloeker Aachen Institute for Automotive Engineering

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Vehicle Intelligence & Automated Driving, Institute for Automotive Engineering,Aachen,Germany,52074

5. Lutz Eckstein Aachen Institute for Automotive Engineering

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Vehicle Intelligence & Automated Driving, Institute for Automotive Engineering,Aachen,Germany,52074