Italy

THDL system include THDL-Det sub-system for text detection and THDL-Rec sub-system for text recognition. The THDL-Det sub-system is an end-to-end text spotter based on the Mask R-CNN instance segmentation framework. The THDL-Rec sub-system adopts a CNN-LSTM-CTC framework to recognize Arabic text lines in videos.

ICPR 2020

TH-DL system for AcTiV2020 Competition

arabic video

text recognition

text detection

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. 


technical paper

Deep image clustering approaches typically use autoencoder
architectures to learn compressed latent representations suitable for clustering tasks. However, they do not effectively regulate the latent space
during training, leading to low performance and diminished applicability
to different datasets. In this paper, we propose a deep clustering model
combining maximum mean discrepancy (MMD) regularization and self-learning clustering optimization to mitigate this problem. Specifically,
we first train the network to improve its image reconstruction ability
by minimizing both reconstruction loss and MMD divergence from a
target distribution. Then, the model gradually learns from its own high-confidence predictions to further optimize the latent distribution. We
validate the network’s performance on different benchmark image sets
using standard clustering metrics, without changing network configuration or adjusting hyper-parameters between datasets. The proposed
model provides top clustering performance across datasets while being
more robust than state-of-the-art methods.

Deep Image Clustering Using Self-Learning Optimization in a Variational Auto-Encoder

To facilitate deep learning based table extraction from unstructured digital files, we disclose a standard Chinese financial dataset from PDF files for table extraction benchmark test, which is diverse, sufficient and comprehensive. We also propose a GCN-based algorithm GFTE as a baseline with a novel idea of integrating all possible types of ground truth together.

GFTE: Graph-based Financial Table Extraction

Relative attributes indicate the strength of a particular attribute between image pairs. We introduce a deep Siamese network with rank SVM loss function, called Deep-RankSVM, that can decide which one of a pair of images has a stronger presence of a specific attribute. The network is trained in an end-to-end fashion to jointly learn the visual features and the ranking function. The trained network for an attribute can predict the relative strength of that attribute in novel images. We demonstrate the effectiveness of our approach against the state-of-the-art methods on four image benchmark datasets: LFW-10, PubFig, UTZap50K-2 and UTZap50K-lexi datasets. Deep-RankSVM surpasses state-of-art in terms of the average accuracy across attributes, on three of the four image benchmark datasets.

Relative Attribute Classification with Deep-RankSVM

In this paper, we propose the research of the recursive division approach to get an explanation for a particular decision of the shallow black box model. The core of the proposed method is the division of the image being classified into separate rectangular parts, followed by the analysis of their influence on the classification result. Such divisions are repeated recursively until the explanation of the classification result is found, or the size of parts is too small. As a result, the pair of images with complement hidden parts is discovered, the first one of which preserves both the most valuable parts and the classification result of the initial image. The second image represents the result of hiding the most valuable parts of an initial image that leads to the different classification for the binary classification problem. Experimental research (applied for Food-5K and concrete crack images datasets) proved that the quality of the proposed method might be close to LIME or even better, while the performance of recursive division is better.

Recursive Division of Image for Explanation of Shallow CNN Models

Deep Neural Networks are becoming the prominet solution
when using machine learning models. However, they suffer of a black-
box effect that renders complicated their inner workings interpretation
and thus the understanding of their successes and failures. Information
visualisation is one way among others to help in their interpretability
and hypothesis deduction. This paper presents a novel way to visualize
a trained DNN in order to depict at the same time its architecture
and its way of treating the classes of a test dataset at the layer level.
This way, it is possible to visually detect where the DNN starts to be
able to discriminate the classes or where it could decrease its separation
ability (and thus detect an oversized network). We have implemented the
approach and validated it using several well-known datasets and networks.
Results show the approach is promising and deserves further studies.

Samples Classification Analysis Across DNN Layers with Fractal Curves

Explaining the behaviors of deep neural networks, usually considered as black boxes, is critical especially when they are now being adopted over diverse aspects of human life. Taking the advantages of interpretable machine learning (interpretable ML), this paper proposes a novel tool called Catastrophic Forgetting Dissector (or CFD) to explain catastrophic forgetting in continual learning settings. We also introduce a new method called Critical Freezing based on the observations of our tool. Experiments on ResNet-50 articulate how catastrophic forgetting happens, particularly showing which components of this famous network are forgetting. Our new continual learning algorithm defeats various recent techniques by a significant margin, proving the capability of the investigation. Critical freezing not only attacks catastrophic forgetting but also exposes explainability.

Explaining How Deep Neural Networks Forget by Deep Visualization

In this talk we present our work on a perplexity-based sparsity definition to derive and visualise layer-wise activation measures. These novel explainable AI strategies reveal a surprising relationship between activation sparsity and overfitting. Our methods reveal novel insights into the contradicting concepts of activation sparsity and network capacity by demonstrating that dense activations can enable discriminative feature learning while efficiently exploiting the capacity of deep models without suffering from overfitting, even when trained excessively.

The Impact of Activation Sparsity on Overfitting in Convolutional Neural Networks

Explanations for algorithmic decision systems can take different forms, they can target different types of users with different goals. One of the main challenges in this area is therefore to devise explanation methods that can accommodate this variety of situations. A first step to address this challenge is to allow explainees to express their needs in the most convenient way, depending on their level of expertise and motivation. In this paper, we present a solution to this problem based on a multi-layered approach allowing users to express their requests for explanations at different levels of abstraction. We illustrate the approach with the application of a proof-of-concept system called IBEX to two case studies.

A Multi-layered Approach for Tailored Black-box Explanations

Machine learning is currently successfully used for addressing several cybersecurity detection and classification tasks. Typically, such detectors are modeled through complex learning algorithms employing a wide variety of features. Although these settings allow achieving considerable performances, gaining insights on the learned knowledge turns out to be a hard task. To address this issue, research efforts on the interpretability of machine learning approaches to cybersecurity tasks is currently rising. In particular, relying on explanations could improve prevention and detection capabilities since they could help human experts to find out the distinctive features that truly characterize malware attacks. In this perspective, Android ransomware represents a serious threat. Leveraging state-of-the-art explanation techniques, we present a first approach that enables the identification of the most influential discriminative features for ransomware characterization. We propose strategies to adopt explanation techniques appropriately and describe ransomware families and their evolution over time. Reported results suggest that our proposal can help cyber threat intelligence teams in the early detection of new ransomware families, and could be applicable to other malware detection systems through the identification of their distinctive features.

Explanation-driven Characterization of Android Ransomware

This paper profiles the recent research work on eXplainable AI (XAI), at the Insight Centre for Data Analytics. This work concentrates on post-hoc ex-planation-by-example solutions to XAI as one approach to explaining black box deep-learning systems. Three different methods of post-hoc ex-planation are outlined for image and time-series datasets: that is, factual, counterfactual, and semi-factual methods). The future landscape for XAI solutions is discussed.

Premium content

TH-DL system for AcTiV2020 Competition

Downloads

Next from ICPR 2020

Deep Image Clustering Using Self-Learning Optimization in a Variational Auto-Encoder

Similar lecture

DUET: Detection Utilizing Enhancement for Text in Scanned or Captured Documents