Abstract: The emergence of transformer models like BERT, GPT-2, GPT-3, RoBERTa, T5 for natural language understanding tasks has opened the floodgates towards solving a wide array of machine learning tasks in other modalities like images, audio, music, sketches and so on. These language models are domain-agnostic and as a result could be applied to 1-D sequences of any kind. However, the key challenge lies in bridging the modality gap so that they could generate strong features beneficial for out-of-domain tasks. This work focuses on leveraging the power of such pre-trained language models and discusses the challenges in predicting challenging handwritten symbols and alphabets.

Abstract: Resting-State fMRI (RS-fMRI) is a brain imaging technique useful for exploring functional connectivity. A major point of interest in RS-fMRI analysis is to isolate connectivity patterns characterising disorders such as for instance ADHD. Such characterisation is usually performed in two steps: first, all connectivity patterns in the data are extracted by means of Independent Component Analysis (ICA); second, standard statistical tests are performed over the extracted patterns to find differences between control and clinical groups. In this work we introduce a novel, single-step, approach for this problem termed Spatial Discriminant ICA. The algorithm can efficiently isolate networks of functional connectivity characterising a clinical group by combining ICA and a new variant of the Fisher’s Linear Discriminant also introduced in this work. As the characterisation is carried out in a single step, it potentially provides for a richer characterisation of inter-class differences. The algorithm is tested using synthetic and real fMRI data, showing promising results in both experiments.

Abstract: This study presents an efficient super-pixel extraction algorithm with major contributions to the state-of-the-art in terms of accuracy and computational complexity. Segmentation accuracy is improved through convexity constrained geodesic distance utilization; while computational efficiency is achieved by replacing complete region processing with boundary adaptation idea. Starting from the uniformly distributed rectangular equal-sized super-pixels, region boundaries are adapted to intensity edges iteratively by assigning boundary pixels to the most similar neighboring super-pixels. At each iteration, super-pixel regions are updated and hence progressively converging to compact pixel groups. Experimental results with state-of-the-art comparisons, validate the performance of the proposed technique in terms of both accuracy and speed.

Abstract: The first requisite of a robot to succeed in social interactions is accurate human localisation, i.e. subject detection and tracking. Later, it is estimated whether an interaction partner seeks attention, for example by interpreting the position and orientation of the body. In computer vision, these cues usually are obtained in colour images, whose qualities are degraded in ill illuminated social scenes. In these scenarios depth sensors offer a richer representation. Therefore, it is important to combine colour and depth information. The
second aspect that plays a fundamental role in the acceptance of social robots is their real-time-ability. Processing colour and depth images is computationally demanding. To overcome this we propose a parallelisation strategy of face detection and tracking based on two different architectures: message passing and shared memory. Our results demonstrate high accuracy in
low computational time, processing nine times more number of frames in a parallel implementation. This provides a real-time social robot interaction.

Abstract: Recognizing visual categories from semantic descriptions is a promising way to extend the capability of a visual classifier beyond the concepts represented in the training data (i.e. seen categories). This problem is addressed by (generalized) zero-shot learning methods (GZSL), which leverage semantic descriptions that connect them to seen categories (e.g. label embedding, attributes). Conventional GZSL are designed mostly for object recognition. In this paper we focus on zero-shot scene recognition, a more challenging setting with hundreds of categories where their differences can be subtle and often localized in certain objects or regions. Conventional GZSL representations are not rich enough to capture these local discriminative differences. Addressing these limitations, we propose a feature generation framework with two novel components: 1) multiple sources of semantic information (i.e. attributes, word embeddings and descriptions), 2) region descriptions that can enhance scene discrimination. To generate synthetic visual features we propose a two-step generative approach, where local descriptions are sampled and used as conditions to generate visual features. The generated features are then aggregated and used together with real features to train a joint classifier. In order to evaluate the proposed method, we introduce a new dataset for zero-shot scene recognition with multi-semantic annotations. Experimental results on the proposed dataset and SUN Attribute dataset illustrate the effectiveness of the proposed method.

Abstract: This article presents an efficient approach to classify a set of corn kernels in contact, which may contain good, or defective kernels along with impurities. The proposed approach consists of two stages, the first one is a next-generation segmentation network, trained by using a set of synthesized images that is applied to divide the given image into a set of individual instances. An ad-hoc lightweight CNN architecture is then proposed to classify each instance into one of three categories (ie good, defective, and impurities). The segmentation network is trained using a strategy that avoids the time-consuming and human-error-prone task of manual data annotation. Regarding the classification stage, the proposed ad-hoc network is designed with only a few sets of layers to result in a lightweight architecture capable of being used in integrated solutions. Experimental results and comparisons with previous approaches showing both the improvement in accuracy and the reduction in time are provided. Finally, the segmentation and classification approach proposed can be easily adapted for use with other cereal types.

Abstract: This paper presents a novel approach for thermal super-resolution based on a fusion prior, low-resolution thermal image and H brightness channel of the corresponding visible spectrum image. The method combines bicubic interpolation of the ×8 scale target image with the brightness component. To enhance the guidance process, the original RGB image is converted to HSV, and the brightness channel is extracted. Bicubic interpolation is then applied to the low-resolution thermal image, resulting in a Bicubic-Brightness channel blend. This luminance-bicubic fusion is used as an input image to help the training process. With this fused image, the cyclic adversarial generative network obtains high-resolution thermal image results. Experimental evaluations show that the proposed approach significantly improves spatial resolution and pixel intensity levels compared to other state-of-the-art techniques, making it a promising method to obtain high-resolution thermal.

Abstract: This paper provides an overview of the DocILE 2023 Competition, its tasks, participant submissions, the competition results and possible future research directions. This first edition of the competition focused on two Information Extraction tasks, Key Information Localization and Extraction (KILE) and Line Item Recognition (LIR). Both of these tasks require detection of pre-defined categories of information in business documents. The second task additionally requires correctly grouping the information into tuples, capturing the structure laid out in the document. The competition used the recently published DocILE dataset and benchmark that stays open to new submissions. The diversity of the participant solutions indicates the potential of the dataset as the submissions included pure Computer Vision, pure Natural Language Processing, as well as multi-modal solutions and utilized all of the parts of the dataset, including the annotated, synthetic and unlabeled subsets.