Self-Supervised Learning for Autonomous Vehicles Perception: A Conciliation Between Analytical and Learning Methods

  title={Self-Supervised Learning for Autonomous Vehicles Perception: A Conciliation Between Analytical and Learning Methods},
  author={Florent Chiaroni and Mohamed-Cherif Rahal and Nicolas Hueber and Fr{\'e}d{\'e}ric Dufaux},
  journal={IEEE Signal Processing Magazine},
The interest in autonomous driving has continuously increased in the last two decades. However, to be adopted, such critical systems need to be safe. Concerning the perception of the ego-vehicle environment, the literature has investigated two different types of methods. On the one hand, traditional analytical methods generally rely on handcrafted designs and features while on the other hand, learning methods aim at designing their own appropriate representation of the observed scene. 

Figures from this paper

On the Road With 16 Neurons: Towards Interpretable and Manipulable Latent Representations for Visual Predictions in Driving Scenarios
A strategy for visual perception in the context of autonomous driving is proposed that uses compact representations that use as few as 16 neural units for each of the two basic driving concepts the authors consider: cars and lanes. Expand
Weakly supervised learning for image classification and potentially moving obstacles analysis
Dans le contexte des applications de perception pour le vehicule a conduite deleguee, l’interet pour les approches d’apprentissage automatique a continuellement augmente pendant cette derniereExpand
Feature-Filter: Detecting Adversarial Examples through Filtering off Recessive Features
  • Hui Liu, Bo Zhao, Yuefeng Peng, Jiabao Guo, Peng Liu
  • Computer Science
  • ArXiv
  • 2021
It is revealed that imperceptibility of the adversarial examples indicates that the perturbations enrich recessive features, yet hardly affect dominant features, and a label-only adversarial detection approach that is referred to as feature-filter is proposed that can real-time detect imperceptible adversarialExamples at high accuracy and few false positives. Expand
Unsupervised Domain Adaption for High-Resolution Coastal Land Cover Mapping with Category-Space Constrained Adversarial Network
A category-space constrained adversarial method to execute category-level adaptive CLCM with high-resolution remotely sensed images and the self-supervised learning approach is also leveraged as an improvement strategy to optimize the result within segmented training. Expand


Learning long-range vision for autonomous off-road driving
Most vision-based approaches to mobile robotics suffer from the limitations imposed by stereo obstacle detection, which is short range and prone to failure. We present a self-supervised learningExpand
Learning long-range vision for autonomous off-road driving
This work presents a self-supervised learning process for long-range vision that is able to accurately classify complex terrain at distances up to the horizon, thus allowing superior strategic planning. Expand
Visual Odometry [Tutorial]
Visual odometry (VO) is the process of estimating the egomotion of an agent (e.g., vehicle, human, and robot) using only the input of a single or If multiple cameras attached to it. ApplicationExpand
Are we ready for autonomous driving? The KITTI vision benchmark suite
The autonomous driving platform is used to develop novel challenging benchmarks for the tasks of stereo, optical flow, visual odometry/SLAM and 3D object detection, revealing that methods ranking high on established datasets such as Middlebury perform below average when being moved outside the laboratory to the real world. Expand
Self-supervised Monocular Road Detection in Desert Terrain
This method for identifying drivable surfaces in difficult unpaved and offroad terrain conditions as encountered in the DARPA Grand Challenge robot race achieves robustness by combining sensor information from a laser range finder, a pose estimation system and a color camera. Expand
Adaptive Road Following using Self-Supervised Learning and Reverse Optical Flow
A road following algorithm that operates in a selfsupervised learning regime, allowing it to adapt to changing road conditions while making no assumptions about the general structure or appearance of the road surface is proposed. Expand
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving
This paper presents an approach to joint classification, detection and semantic segmentation using a unified architecture where the encoder is shared amongst the three tasks, and performs extremely well in the challenging KITTI dataset. Expand
Deep MANTA: A Coarse-to-Fine Many-Task Network for Joint 2D and 3D Vehicle Analysis from Monocular Image
This paper presents a novel approach, called Deep MANTA (Deep Many-Tasks), for many-task vehicle analysis from a given image, based on a new coarse-to-fine object proposal that boosts the vehicle detection. Expand
Unsupervised Learning of Depth and Ego-Motion from Video
Empirical evaluation demonstrates the effectiveness of the unsupervised learning framework for monocular depth performs comparably with supervised methods that use either ground-truth pose or depth for training, and pose estimation performs favorably compared to established SLAM systems under comparable input settings. Expand
Traversability analysis using terrain mapping and online-trained Terrain type classifier
This paper presents a path detection method that mixes together 3D mapping and visual classification, trying to learn, in real time, the actual road characteristics, using an on-line learning of visual characteristics to feedback a terrain classifier. Expand