“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

A scheme is developed for classifying the types of motion perceived by a humanlike robot. It is assumed that the robot receives visual images of the scene using a perspective system model. Equations, theorems, concepts, clues, etc., relating the objects, their positions, and their motion to their images on the focal plane are presented. >

http://ieeexplore.ieee.org/iel2/815/3148/00100651.pdf

Robot vision

Robust detection and tracking of objects is crucial for the deployment of autonomous vehicle technology. Image based benchmark datasets have driven development in computer vision tasks such as object detection, tracking and segmentation of agents in the environment. Most autonomous vehicles, however, carry a combination of cameras and range sensors such as lidar and radar. As machine learning based methods for detection and tracking become more prevalent, there is a need to train and evaluate such methods on datasets containing range sensor data along with images. In this work we present nuTonomy scenes (nuScenes), the first dataset to carry the full autonomous vehicle sensor suite: 6 cameras, 5 radars and 1 lidar, all with full 360 degree field of view. nuScenes comprises 1000 scenes, each 20s long and fully annotated with 3D bounding boxes for 23 classes and 8 attributes. It has 7x as many annotations and 100x as many images as the pioneering KITTI dataset. We define novel 3D detection and tracking metrics. We also provide careful dataset analysis as well as baselines for lidar and image based detection and tracking. Data, development kit and more information are available online.

nuScenes: A multimodal dataset for autonomous driving

/pdf/nuscenes-a-multimodal-dataset-for-autonomous-driving-3znmpkqyuu.pdf

nuScenes: A Multimodal Dataset for Autonomous Driving

Object detection, as of one the most fundamental and challenging problems in computer vision, has received great attention in recent years. Its development in the past two decades can be regarded as an epitome of computer vision history. If we think of today's object detection as a technical aesthetics under the power of deep learning, then turning back the clock 20 years we would witness the wisdom of cold weapon era. This paper extensively reviews 400+ papers of object detection in the light of its technical evolution, spanning over a quarter-century's time (from the 1990s to 2019). A number of topics have been covered in this paper, including the milestone detectors in history, detection datasets, metrics, fundamental building blocks of the detection system, speed up techniques, and the recent state of the art detection methods. This paper also reviews some important detection applications, such as pedestrian detection, face detection, text detection, etc, and makes an in-deep analysis of their challenges as well as technical improvements in recent years.

Object Detection in 20 Years: A Survey

Convolutional neural networks are designed for dense data, but vision data is often sparse (stereo depth, point clouds, pen stroke, etc.). We present a method to handle sparse depth data with optional dense RGB, and accomplish depth completion and semantic segmentation changing only the last layer. Our proposal efficiently learns sparse features without the need of an additional validity mask. We show how to ensure network robustness to varying input sparsities. Our method even works with densities as low as 0.8% (8 layer lidar), and outperforms all published state-of-the-art on the Kitti depth completion benchmark.

Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentation

This paper introduces a new real-time traffic light recognition system for on-vehicle camera applications. This approach has been tested with good results in urban scenes. Thanks to the use of our generic “Adaptive Templates” it would be possible to recognize different kinds of traffic lights from various countries.

Real time visual traffic lights recognition based on Spot Light Detection and adaptive traffic lights templates

This paper introduces a new real-time traffic light recognition system for on-vehicle camera applications. This approach has been tested with good results in urban scenes. Thanks to the use of our generic “Adaptive Templates” it would be possible to recognize different kinds of traffic lights from various countries. This approach is mainly based on a spot detection algorithm therefore able to detect lights from a high distance with the main advantage of being not so sensitive to motion blur and illumination variations. The detected spots together with other shape analysis form strong hypothesis we feed our Adaptive Templates Matcher with. Even though it is still in progress, our system was validated in real conditions in our prototype vehicle and also using registered video sequences. The authors noticed a high rate of correctly recognized traffic lights and very few false alarms. Processing is performed in real-time on 640x480 images using a 2.9GHz single core desktop computer.

Real Time Visual Traffic Lights Recognition Based on Spot Light Detection and Adaptive Traffic Lights Templates

We present research using the latest reinforcement learning algorithm for end-to-end driving without any mediated perception (object recognition, scene understanding). The newly proposed reward and learning strategies lead together to faster convergence and more robust driving using only RGB image from a forward facing camera. An Asynchronous Actor Critic (A3C) framework is used to learn the car control in a physically and graphically realistic rally game, with the agents evolving simultaneously on tracks with a variety of road structures (turns, hills), graphics (seasons, location) and physics (road adherence). A thorough evaluation is conducted and generalization is proven on unseen tracks and using legal speed limits. Open loop tests on real sequences of images show some domain adaption capability of our method.

End-to-End Race Driving with Deep Reinforcement Learning

Unsupervised Domain Adaptation (UDA) is crucial to tackle the lack of annotations in a new domain. There are many multi-modal datasets, but most UDA approaches are uni-modal. In this work, we explore how to learn from multi-modality and propose cross-modal UDA (xMUDA) where we assume the presence of 2D images and 3D point clouds for 3D semantic segmentation. This is challenging as the two input spaces are heterogeneous and can be impacted differently by domain shift. In xMUDA, modalities learn from each other through mutual mimicking, disentangled from the segmentation objective, to prevent the stronger modality from adopting false predictions from the weaker one. We evaluate on new UDA scenarios including day-to-night, country-to-country and dataset-to-dataset, leveraging recent autonomous driving datasets. xMUDA brings large improvements over uni-modal UDA on all tested scenarios, and is complementary to state-of-the-art UDA techniques. Code is available at https://github.com/valeoai/xmuda.

/pdf/xmuda-cross-modal-unsupervised-domain-adaptation-for-3d-56fu5tfqk8.pdf

Raoul de Charette

Papers

Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentation

Real time visual traffic lights recognition based on Spot Light Detection and adaptive traffic lights templates

Real Time Visual Traffic Lights Recognition Based on Spot Light Detection and Adaptive Traffic Lights Templates

End-to-End Race Driving with Deep Reinforcement Learning

xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation