Italy

Late fusion (LF) architectures are common approaches in classification tasks that consist of several feature subspaces, e.g. multi-modal sensor data, such as a combination of audio, video, and physiological signals. In the current study, we propose a trainable sample-specific LF architecture that combines classification and weighting models (CMs, WMs). The idea of our approach is to train each WM in combination with a set of CM-specific meta labels.

ICPR 2020

Using Meta Labels for the Training of Weighting Models in a Sample-Specific Late Fusion Classification Architecture

trainable aggregation rule

information fusion

classification

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

Online fraud often involves identity theft. Since most security measures are weak or can be spoofed, we investigate a more nuanced and less explored avenue: behavioral biometrics via handwriting movements. This kind of data can be used to verify whether a user is operating a device or a computer application, so it is important to distinguish between human and machine-generated movements reliably. For this purpose, we study handwritten symbols (isolated characters, digits, gestures, and signatures) produced by humans and machines, and compare and contrast several deep learning models. We find that if symbols are presented as static images, they can fool state-of-the-art classifiers (near 75% accuracy in the best case) but can be distinguished with remarkable accuracy if they are presented as temporal sequences (95% accuracy in the average case). We conclude that an accurate detection of fake movements has more to do with how users write, rather than what they write. Our work has implications for computerized systems that need to authenticate or verify legitimate human users, and provides an additional layer of security to keep attackers at bay.

Human or Machine? It Is Not What You Write, But How You Write It

We present an approach to detect basic movements of cyclists in real-world traffic situations based on image sequences, optical flow (OF) sequences, and past positions using a multi-stream 3D convolutional neural network (3D-ConvNet) architecture. To resolve occlusions of cyclists by other traffic participants or road structures, we use a wide-angle stereo camera system mounted at a heavily frequented public intersection. We created a large dataset consisting of 1,639
video sequences containing cyclists, recorded in real-world traffic, resulting in over 1.1 million samples. Through modeling the cyclists’ behavior by a state machine of basic cyclist movements, our approach takes every situation into account and is not
limited to certain scenarios. We compare our method to an approach solely based on position sequences. Both methods are evaluated taking into account frame-wise and scene wise classification results of basic movements, and detection times of basic movement transitions, where our approach outperforms the position-based approach by producing more reliable detections with shorter detection times. Our code and parts of our dataset are made publicly available.

Image Sequence Based Cyclist Action Recognition Using Multi-Stream 3D Convolution

Object detection in images has reached unprecedented performances. The state-of-the-art methods rely on deep architectures that extract salient features and predict bounding boxes enclosing the objects of interest. These methods essentially run on RGB images. However, the RGB images are often compressed by the acquisition devices for storage purpose and transfer efficiency. Hence, their decompression is required for object detectors. To gain in efficiency, this paper proposes to take advantage of the compressed representation of images to carry out object detection usable in constrained resources conditions.
Specifically, we focus on JPEG images and propose a thorough analysis of detection architectures newly designed in regard of the peculiarities of the JPEG norm. This leads to a x1.7 speed up in comparison with a standard RGB-based architecture, while only reducing the detection performance by 5.5%. Additionally, our empirical findings demonstrate that only part of the compressed JPEG information, namely the luminance component, may be required to match detection accuracy of the full input methods.

Object Detection in the DCT Domain: is Luminance the Solution?


The demands of automated shipping address recognition and verification have increased to handle a large number of packages and to save costs associated with misdelivery. A previous study proposed a deep learning system where the shipping address is recognized and verified based on a camera image capturing the shipping address and barcode area. Because the system performance depends on the input image quality, inspection of input image quality is necessary for image preprocessing. In this paper, we propose an input image quality verification method combining global and local features. Object detection and scale-invariant feature transform in different feature spaces are developed to extract global and local features from several independent convolutional neural networks. The conditions of shipping label images are classified by fully connected fusion layers with concatenated global and local features. The experimental results regarding real captured and generated images show that the proposed method achieves better performance than other methods. These results are expected to improve the shipping address recognition and verification system by applying different image preprocessing steps based on the classified conditions.

Fusion of Global-Local Features for Image Quality inspection of Shipping Label


Most of existing superpixel methods are designed to segment standard planar images as pre-processing for computer vision pipelines. Nevertheless, the 
increasing number of applications based on wide angle capture devices, mainly generating 360o spherical images, have enforced the need for dedicated superpixel approaches. In this paper, we introduce a new superpixel method for spherical images called SphSPS (for Spherical Shortest Path-based Superpixels). Our approach respects the spherical geometry and generalizes the notion of shortest path between a pixel and a superpixel center on the 3D spherical acquisition space. To relevantly evaluate this last aspect in the spherical space, we also generalize a planar global regularity metric. Finally, the proposed SphSPS method obtains significantly better performances than both planar and spherical recent superpixel approaches on the reference 360o spherical panorama segmentation dataset.

Generalized Shortest Path-based Superpixels for Accurate Segmentation of Spherical Images

Spherical cameras and the latest image processing techniques open up new horizons. In particular, methods based on Convolutional Neural Networks (CNNs) now give excellent results for optical flow estimation on perspective images. However, these approaches are highly dependent on their architectures and training datasets. This paper proposes to benefit from years of improvement in perspective images optical flow estimation and to apply it to omnidirectional ones without training on new datasets. Our network, OmniFlowNet, is built on a CNN specialized in perspective images. Its convolution operation is adapted to be consistent with the equirectangular projection. Tested on spherical datasets created with Blender and several equirectangular videos realized from real indoor and outdoor scenes, OmniFlowNet shows better performance than its original network without extra training.

OmniFlowNet: a Perspective Neural Network Adaptation forOptical Flow Estimation in Omnidirectional Images

Multi-Task learning (MTL) provides state-of-the-art results
in many applications of computer vision and natural
language processing. In contrast to single-task learning
(STL), MTL allows for leveraging knowledge between related
tasks improving prediction results on all tasks. However,
there is a limited number of comparative studies applied to
MTL architectures for regression and time series problems
taking recent advances of MTL into account. An intriguing,
non-linear time-series problem are day ahead forecasts of
the expected power generation for renewable power plants.
Therefore, the main contribution of this article is a
comparative study of the following recent and relevant MTL
architectures: Hard-parameter sharing, cross-stitch
network, and sluice network (SN). They are compared to a
multi-layer peceptron (MLP) model of similar size in an STL
setting. As a additional contribution, we provide a simple,
yet practical approach to model task specific information
through an embedding layer in an MLP, referred to as task
embedding. Further, we contribute a new MTL architecture
named emerging relation network (ERN), which can be
considered as an extension of the SN. 
For a solar power dataset, the task embedding achieves the
best mean improvement with 8.2%. For two wind and one
additional solar dataset, the ERN is the best MTL
architecture with improvements up to 11.3%.

Emerging Relation Network and Task Embedding for Multi-Task Regression Problems

Different techniques have emerged in the deep learning scenario, such as Convolutional Neural Networks, Deep Belief Networks, and Long Short-Term Memory Networks, to cite a few. In lockstep, regularization methods, which aim to prevent overfitting by penalizing the weight connections, or turning off some units, have been widely studied either. In this paper, we present a novel approach called MaxDropout, a regularizer for deep neural network models that works in a supervised fashion by removing (shutting off) the prominent neurons (i.e., most active) in each hidden layer. The model forces fewer activated units to learn more representative information, thus providing sparsity. Regarding the experiments, we show that it is possible to improve existing neural networks and provide better results in neural networks when Dropout is replaced by MaxDropout. The proposed method was evaluated in image classification, achieving comparable results to existing regularizers, such as Cutout and RandomErasing, also improving the accuracy of neural networks that uses Dropout by replacing the existing layer by MaxDropout.

MaxDropout: Deep Neural Network Regularization Based on Maximum Output Values

Few-shot learning is a challenging problem that has attracted more and more attention recently since abundant training samples are difficult to obtain in practical applications. Meta-learning has been proposed to address this issue, which focuses on quickly adapting a predictor as a base-learner to new tasks, given limited labeled samples. However, a critical challenge for meta-learning is the representation deficiency since it is hard to discover common information from a small number of training samples or even one, as is the representation of key features from such little information. As a result, a meta-learner cannot be trained well in a high-dimensional parameter space to generalize to new tasks. Existing methods mostly resort to extracting less expressive features so as to avoid the representation deficiency. Aiming at learning better representations, we propose a metalearning approach with complemented representations network (MCRNet) for few-shot image classification. In particular, we embed a latent space, where latent codes are reconstructed with extra representation information to complement the representation deficiency. Furthermore, the latent space is established with variational inference, collaborating well with different baselearners, and can be extended to other models. Finally, our endto-end framework achieves the state-of the-art performance in image classification on three standard few-shot learning datasets.

Complementing Representation Deficiency in Few-shot Image Classification: A Meta-Learning Approach

Updating the tracker model with adverse bounding box predictions adds an unavoidable bias term to the learning. This bias term, which we refer to as model decay, offsets the learning and causes tracking drift. While its adverse affect
might not be visible in short-term tracking, accumulation of this bias over a long-term can eventually lead to a permanent loss of the target. In this paper, we look at the problem of model bias from a mathematical perspective. Further, we briefly examine the effect of various sources of tracking error on model decay, using a correlation filter (ECO) and a Siamese (SINT) tracker. Based on observations and
insights, we propose simple additions that help to reduce model decay in long-term tracking. The proposed tracker is evaluated on four long-term and one short term tracking benchmarks, demonstrating superior accuracy and robustness, even in 30 minute long videos.

Premium content

Using Meta Labels for the Training of Weighting Models in a Sample-Specific Late Fusion Classification Architecture

Downloads

Next from ICPR 2020

Human or Machine? It Is Not What You Write, But How You Write It

Similar lecture

Beyond cross-entropy: learning highly separable feature distributions for robust and accurate classification