Italy

Autonomous driving is challenging in adverse road and weather conditions in which there might not be lane lines, the road might be covered in snow and the visibility might be poor. We extend the previous work on end-to-end learning for autonomous steering to operate in these adverse real-life conditions with multimodal data. We collected 28 hours of driving data in several road and weather conditions and trained convolutional neural networks to predict the car steering wheel angle from front-facing color camera images and lidar range and reflectance data. We compared the CNN model performances based on the different modalities and our results show that the lidar modality improves the performances of different multimodal sensor-fusion models. We also performed on-road tests with different models and they support this observation.

ICPR 2020

Multimodal End-to-End Learning for Autonomous Steering in Adverse Road and Weather Conditions

adverse weather conditions

autonomous driving

convolutional neural networks

deep learning

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


poster

Visual object tracking is of great interest in many applications, as it preserves the identity of an object throughout a video. However, while real applications demand systems capable of real-time-tracking multiple objects, multi-object tracking solutions usually follow the tracking-by-detection paradigm, thus they depend on running a costly detector in each frame, and they do not allow the tracking of arbitrary objects, i.e., they require training for specific classes. In response to this need, this work presents the architecture of SiamMT, a system capable of efficiently applying individual visual tracking techniques to multiple objects in real-time. This makes it the first deep-learning-based arbitrary multi-object tracker. To achieve this, we propose global frame features extraction by using a fully-convolutional neural network, followed by the cropping and resizing of the different object search areas. The final similarity operation between these search areas and the target exemplars is carried out with an optimized pairwise cross-correlation. These novelties allow the system to track multiple targets in a scalable manner, achieving 25 fps with 60 simultaneous objects for VGA videos and 40 objects for HD720 videos, all with a tracking quality similar to SiamFC.

SiamMT: Real-Time Arbitrary Multi-Object Tracking

We propose a benchmark for automated cooperative maneuvers. The targeted cooperative maneuver is the vehicle lane merge where a vehicle on the acceleration
lane merges into the traffic of a motorway. The benchmark enables the evaluation of vehicle localization approaches as well as the study of cooperative maneuvers. It consists of temporally synchronized multi-view video streams, highly accurate camera calibration, and ground truth vehicle descriptions, including position, heading, speed, and shape. For benchmark generation, the lane merge maneuver is performed by human drivers on a test track, resulting in 85 lane merge data sets with various traffic situations and video recording conditions.

Vehicle Lane Merge Visual Benchmark


In this paper, we present the PS^2-Net - a locally and globally aware deep learning framework for semantic segmentation on 3D scene-level point clouds. In order to deeply incorporate local structures and global context to support 3D scene segmentation, our network is built on four repeatedly stacked encoders, where each encoder has two basic components: EdgeConv that captures local structures and NetVLAD that models global context. Different from existing state-of-the-art methods for point-based scene semantic segmentation that either violate or do not achieve permutation invariance, our PS^2-Net is designed to be permutation invariant which is an essential property of any deep network used to process unordered point clouds. We further provide theoretical proof to guarantee the permutation invariance property of our network. We perform extensive experiments on two large-scale 3D indoor scene datasets and demonstrate that our PS^2-Net is able to achieve state-of-the-art performances as compared to existing approaches.

PS^2-Net: A Locally and Globally Aware Network for Point-Based Semantic Segmentation


In this paper, we propose the application of con- ditional generative adversarial networks to solve various phase retrieval problems. We show that including knowledge of the measurement process at training time leads to an optimization at test time that is more robust to initialization than existing approaches involving generative models. In addition, conditioning the generator network on the measurements enables us to achieve much more detailed results. We empirically demonstrate that these advantages provide meaningful solutions to the Fourier and the compressive phase retrieval problem and that our method outperforms well-established projection-based methods as well as existing methods that are based on neural networks. Like other deep learning methods, our approach is robust to noise and can therefore be useful for real-world applications.

Phase Retrieval Using Conditional Generative Adversarial Networks

We present a method to generate a video sequence given a single image. Because items in an image can be animated in arbitrarily many different ways, we introduce as control signal a sequence of motion strokes. Such control signal can also be automatically transferred from other videos, e.g., via bounding box tracking. Each motion stroke provides the direction to the moving object in the input image and we aim to train a network to generate an animation following a sequence of such directions. To address this task we design a novel recurrent architecture, which can be trained easily and effectively thanks to an explicit separation of past, future and current states. As we demonstrate in the experiments, our proposed architecture
is capable of generating an arbitrary number of frames from a single image and a sequence of motion strokes. Key components of our architecture are an autoencoding constraint to ensure consistency with the past and a generative adversarial scheme to ensure that images look realistic and are temporally smooth. We demonstrate the effectiveness of our approach on the MNIST, KTH, Human3.6M, Push and Weizmann datasets.

Learning to Take Directions One Step at a Time


Image denoising is an essential part of many image processing and computer vision tasks due to inevitable noise corruption during image acquisition. Traditionally, many researchers have investigated image priors for the denoising, within the Bayesian perspective based on image properties and statistics. Recently, deep convolutional neural networks (CNNs) have shown great success in image denoising by incorporating large-scale synthetic datasets. However, they both have pros and cons. While the deep CNNs are powerful for removing the noise with known statistics, they tend to lack flexibility and practicality for the blind and real-world noise. Moreover, they cannot easily employ explicit priors. On the other hand, traditional non-learning methods can involve explicit image priors, but they require considerable computation time and cannot exploit large-scale external datasets. In this paper, we present a CNN-based method that leverages the advantages of both methods based on the Bayesian perspective. Concretely, we divide the blind image denoising problem into sub-problems and conquer each inference problem separately. As the CNN is a powerful tool for inference, our method is rooted in CNNs and propose a novel design of network for efficient inference. With our proposed method, we can successfully remove blind and real-world noise, with a moderate number of parameters of universal CNN.

Deep Universal Blind Image Denoising


This work introduces the one-shot learning paradigm in the computational bioacoustics domain. Even though, most of the related literature assumes availability of data characterizing the entire class dictionary of the problem at hand, that is rarely true as a habitat’s species composition is only known up to a certain extent. Thus, the problem needs to be addressed by methodologies able to cope with non-stationarity. To this end, we propose a framework able to detect changes in the class dictionary and incorporate new classes on the fly. We design an one-shot learning architecture composed of a Siamese Neural Network operating in the logMel spectrogram space. We extensively examine the proposed approach on two datasets of various bird species using suitable figures of merit. Interestingly, such a learning scheme exhibits state of the art performance, while taking into account extreme non-stationarity cases.

One-shot learning for acoustic identification of bird species in non-stationary environments

Forecasting human actions and motion trajectories address the problem of predicting what a person is going to do next and how they will perform it. This is crucial in a wide range of applications, such as assisted living and future co-robotic settings. We propose to simultaneously learn actions and action-related human motion dynamics while existing works perform them independently. This paper presents a method to jointly forecast categories of human action and skeletal joint pose, allowing the two tasks to reinforce each other. As a result, our system can predict future actions and the motion trajectories that will result. To achieve this, we define a task of joint action classification and pose regression. We employ a sequence to sequence encoder-decoder model combined with multi-task learning to forecast future actions and poses progressively before the action happens. Experimental results on two public datasets, IkeaDB and OAD, demonstrate the effectiveness of the proposed method.

What and How? Jointly Forecasting Human Action and Pose


Scene text recognition has become a hot topic in recent years due to its booming real-life applications. Attention-based encoder-decoder framework has become one of the most popular frameworks especially in the irregular text scenario. However, the “attention drift” problem hinders the recognition performance for most existing attention-based scene text recognition methods. To solve this problem, we propose an auxiliary supervision branch along with the attention-based encoder-decoder framework. A new loss function is designed to refine the feature map and to help the attention region align the target character area. Compared with existing attention rectification mechanisms, our method does not require character-level annotations or introduce any additional trainable parameter. Furthermore, our method can improve the performance for both RNN-Attention and Scaled Dot-Product Attention. The experiment results on various benchmarks have demonstrated that the proposed approach outperforms the state-of-the-art methods in both regular and irregular text recognition scenarios.

Weakly Supervised Attention Rectification for Scene Text Recognition


Focusing on handwritten digit string recognition (HDSR), we propose an anchor-free object detector called ChipNet, where a novel encoding method is designed. The input image is divided into columns, and then these columns are encoded by the ground truth. The adjacent columns are responsible for detecting the same target so that it can well address the class-imbalanced problem meanwhile reducing the network computation. ChipNet is composed of convolutional and bidirectional long short term memory networks. Different from the typical detectors, it doesn't use region proposals, anchors or regions of interest pooling. Hence, it can overcome the shortages of anchor-based and dense detectors in HDSR. The experiments are implemented on the synthetic digit strings, the CVL HDS database, and the ORAND-CAR-A & B databases. The high accuracies, which surpass the reported results by a large margin (up to 6.62%), are achieved. Furthermore, it gets 219 FPS speed on 160*32 px resolution images when using a Tesla P100 GPU. The results also show that ChipNet can handle touching, connecting and arbitrary length digit strings, and the obtained accuracies in HDSR are as high as the ones in single handwritten digit recognition.

Downloads

Next from ICPR 2020

SiamMT: Real-Time Arbitrary Multi-Object Tracking

Similar lecture

MaxDropout: Deep Neural Network Regularization Based on Maximum Output Values