Italy


Given an unknown audio source, the estimation of time differences-of-arrivals (TDOAs) can be efficiently and robustly solved using blind channel identification and exploiting the cross-correlation identity (CCI). Prior “blind” works have improved the estimate of TDOAs by means of different algorithmic solutions and optimization strategies, while always sticking to the case N = 2 microphones. But what if we can obtain a direct improvement in performance by just increasing N? In this paper we try to investigate this direction, showing that, despite the arguable simplicity, this is capable of (sharply) improving upon state-of-the-art blind channel identification methods based on CCI, without modifying the computational pipeline. Inspired by our results, we seek to warm up the community and the practitioners by paving the way (with two concrete, yet preliminary, examples) towards joint approaches in which advances in the optimization are combined with an increased number of microphones, in order to achieve further improvements.

ICPR 2020

Are Multiple Cross-Correlation Identities better than just Two? Improving the Estimate of Time Differences-of-Arrivals from Blind Audio Signals

tdoa estimation

incremental and ensembling approaches

blind channel identification

acoustic impulse response

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

One Class Classification(OCC) is an Extreme case in classification where knowledge of the classifier is limited to only a single class. Given training samples from a class, the one class classifier is expected to reject samples from any outside class. we learn generative features using the one-class data with a generative framework. We augment the learned features with the corresponding reconstruction errors to obtain augmented features. Then, we qualitatively identify a suitable feature distribution that reduces the redundancy in the chosen classifier space. Finally, we force the augmented features to take the form of this distribution using an adversarial framework.

A Joint Representation Learning and Feature Modeling Approach for One-class Recognition


Object proposal technique with dense anchoring scheme for scene text detection were applied frequently to achieve high recall. It results in the significant improvement in accuracy but waste of computational searching, regression and classification. In this paper, we propose an anchor selection-based region proposal network (AS-RPN) using effective selected anchors instead of dense anchors to extract text proposals. The center, scales, aspect ratios and orientations of anchors are learnable instead of fixing, which leads to high recall and greatly reduced numbers of anchors. By replacing the anchor-based RPN in Faster RCNN, the AS-RPN-based Faster RCNN can achieve comparable performance with previous state-of-the-art text detecting approaches on standard benchmarks, including COCOText, ICDAR2013, ICDAR2015 and MSRA-TD500 when using single-scale and single model (ResNet50) testing only.

Scene Text Detection with Selected Anchors

This paper introduces an approach for jointly addressing semantic segmentation (SS) and instance segmentation (IS) of 3D point clouds. Two novel modules are designed to model the interplay between SS and IS. Specifically, we develop an Instance Separation Module that supplements the positioninvariance semantic feature with the instance-specific centroid position to help separate different instances. To fuse the semantic information within a single instance, an attention-based Semantic
Fusion Module is proposed to encode attention maps in the instance embedding space, which are applied to fuse semantic information in the semantic feature space. The proposed method is thoroughly evaluated on the S3DIS dataset. Compared with the state-of-the-art method ASIS, our approach achieves significant improvements across all evaluation metrics in both IS and SS.


Joint Semantic-Instance Segmentation of 3D Point Clouds: Instance Separation and Semantic Fusion


In this work, we aim to estimate 3D bounding boxes by voting to object centers and then groups and aggregates the votes to generate 3D box proposals and semantic classes of objects. However, due to the sparse and unstructured nature of the point clouds, we face some challenges when directly predicting the bounding box from the vote feature: the sparse vote feature may lack some necessary semantic and context information. To address the challenges, we propose a vote feature enhancement
network that aims to encode semantic-aware information and aggravate global context for the vote feature. Specifically, we learn the point-wise semantic information and supplement it to the vote feature, and we also encode the pairwise relations to collect the global context. Experiments on two large datasets of real 3D scans, ScanNet and SUN RGB-D, demonstrate that our method can achieve excellent 3D detection results.

Enhanced Vote Network for 3D Object Detection in Point Clouds


Local binary pattern (LBP) based descriptors have shown effectiveness for texture classification. However, most of them encode the intensity relationships between neighboring pixels and a central pixel into binary forms, thereby failing to capture the complete ordering information among neighbors. Several methods have explored intensity order information for feature description, but they do not address the grayscale-inversion problem. In this paper, we propose an image descriptor called local grouped invariant order pattern (LGIOP) for grayscale-inversion and rotation invariant texture classification. Our LGIOP is a histogram representation which jointly encodes neighboring order information and central pixels. In particular, two new order encoding methods, i.e., intensity order encoding and distance order encoding, are proposed to describe the neighboring relationships. These two order encoding methods are not only complementary but also invariant to grayscale-inversion and rotation changes. Experiments for texture classification demonstrate that the proposed LGIOP descriptor is robust to (linear or nonlinear) grayscale inversion and image rotation.

Local Grouped Invariant Order Pattern for Grayscale-Inversion and Rotation Invariant Texture Classification

Robust Mask R-CNN (Mask Regional Convolutional Neural Network) methods are proposed and tested for automatic detection of cracks on structures or their components that may be damaged during extreme events, such as earthquakes. We curated a new dataset with 2,021 labeled images for training and validation and aimed to find end-to-end deep neural networks for crack detection in the field. With data augmentation and parameters fine-tuning, Path Aggregation Network (PANet) with spatial attention mechanisms and High-resolution Network (HRNet) are introduced into Mask R-CNNs. The tests on three public datasets with low- or high-resolution images demonstrate that the proposed methods can achieve a big improvement over alternative networks, so the proposed method may be sufficient for crack detection for a variety of scales in real applications.

End-to-end Deep Learning Methods for Automated Damage Detection in Extreme Events at Various Scales



A fusion of LiDAR and cameras have been widely used in many robotics applications such as classification, segmentation, object detection, and autonomous driving. It is essential that the LiDAR sensor can measure distances accurately, which is a good complement to the cameras. Hence, calibrating sensors before deployment is a mandatory step. The conventional methods include checkerboards, specific patterns, or human labeling, which is trivial and human-labor extensive if we do the same calibration process every time. The main purpose of this research work is to build a deep neural network that is capable of automatically finding the geometric transformation between LiDAR and cameras. The results show that our model manages to find the transformations from randomly sampled artificial errors. Besides, our work is open-sourced for the community to fully utilize the advances of the methodology for developing more the approach, initiating collaboration, and innovation in the topic.

NetCalib: A Novel Approach for LiDAR-Camera Auto-calibration Based on Deep Learning

This paper proposes two novel convolutional neural networks for 3D object classification, tailored to process point clouds that are composed of planar slices (profiles). In particular, the application that we are targeting is the classification of vehicles by scanning them along planes perpendicular to the driving direction, within the context of Electronic Toll Collection. Depending on sensors configurations, the distance between slices can be measured or not, thus resulting in two types of point
clouds, namely metric and non-metric. In the latter case, two coordinates are indeed metric but the third one is merely a temporal index. Our networks, named SliceNets, extract metric information from the spatial coordinates and neighborhood information from the third one (either metric or temporal), thus being able to handle both types of point clouds. Experiments on two datasets collected in the field show the effectiveness of our networks in comparison with state-of-the-art ones.

Vehicle Classification from Profile Measures

Short-range motion features and long-range dependencies are two complementary and vital cues for action recognition in videos, but it remains unclear how to efficiently and effectively extract these two features. In this paper, we propose a novel network to capture these two features in a unified 2D framework. Specifically, we first construct a Short-range Temporal Interchange (STI) block, which contains a Channels-wise Temporal Interchange (CTI) module for encoding short-range motion features. Then a Graph-based Regional Interchange (GRI) module is built to present long-range dependencies using graph convolution. Finally, we replace original bottleneck blocks in the ResNet with STI blocks and insert several GRI modules between STI blocks, to form a Multi-range Feature Interchange (MFI) Network. Practically, extensive experiments are conducted on three action recognition datasets (i.e., Something-Something V1, HMDB51, and UCF101), which demonstrate that the proposed MFI network achieves impressive results with very limited

MFI: Multi-range Feature Interchange for Video Action Recognition


The analysis of axillary and supraclavicular lymph nodes is a primary prognostic factor for the staging of breast cancer. However, due to the size of lymph nodes and the low resolution of PET data, their segmentation is challenging. We investigate the relevance of considering axillary and supraclavicular lymph node segmentation from PET/CT images by coupling Convolutional Neural Networks (CNNs) and Component-Trees (C-Trees). Building upon the U-Net architecture, we propose a framework that couples a multi-modal U-Net fed with PET and CT, coupled with a hierarchical model obtained from the PET that provides additional high-level region-based features as input channels. Our working hypotheses are twofold. First, we take advantage of both anatomical information from CT for detecting the nodes, and functional information from PET for detecting the pathological ones. Second, we consider region-based attributes extracted from C-Tree analysis of 3D PET/CT images to improve the CNN segmentation. We carried out experiments on a dataset of 240 pathological lymph nodes from 52 patients scans, and compared our outputs with human expert-defined ground-truth, leading to promising results.

Premium content

Are Multiple Cross-Correlation Identities better than just Two? Improving the Estimate of Time Differences-of-Arrivals from Blind Audio Signals

Downloads

Next from ICPR 2020

A Joint Representation Learning and Feature Modeling Approach for One-class Recognition

Similar lecture

3D Pots Configuration System by Optimizing over Geometric Constraints