Carlos Francisco Moreno‐García

Carlos Francisco Moreno-Garcia, Francesc Serratosa

Image registration is a research field in which images must be compared and aligned independently of the point of view or camera characteristics. In some applications (such as forensic biometrics, satellite photography or outdoor scene identification) classical image registration systems fail due to one of the images compared represents a tiny piece of the other image. For instance, in forensics palmprint recognition, it is usual to find only a small piece of the palmprint, but in the database, the whole palmprint has been enrolled. The main reason of the poor behaviour of classical image registration methods is the gap between the amounts of salient points of both images, which is related to the number of points to be considered as outliers. Usually, the difficulty of finding a good match increases when the image that represents the tiny part of the scene has been drastically rotated. Again, in the case of palmprint forensics, it is difficult to decide a priori the orientation of the found tiny palmprint image. We present a rotation invariant registration method that explicitly considers that the image to be matched is a small piece of a larger image. We have experimentally validated our method in two different scenarios; palmprint identification and outdoor image registration.

Joseph De Mathia, Carlos Francisco Moreno-García

In an era where wearable technology is reshaping applications, Scene Text Detection and Recognition (STDR) becomes a straightforward choice through the lens of egocentric vision. Leveraging Meta's Project Aria smart glasses, this paper investigates how environmental variables, such as lighting, distance, and resolution, affect the performance of state-of-the-art STDR algorithms in real-world scenarios. We introduce a novel, custom-built dataset captured under controlled conditions and evaluate two OCR pipelines: EAST with CRNN, and EAST with PyTesseract. Our findings reveal that resolution and distance significantly influence recognition accuracy, while lighting plays a less predictable role. Notably, image upscaling emerged as a key pre-processing technique, reducing Character Error Rate (CER) from 0.65 to 0.48. We further demonstrate the potential of integrating eye-gaze tracking to optimise processing efficiency by focusing on user attention zones. This work not only benchmarks STDR performance under realistic conditions but also lays the groundwork for adaptive, user-aware AR systems. Our contributions aim to inspire future research in robust, context-sensitive text recognition for assistive and research-oriented applications, such as asset inspection and nutrition analysis. The code is available at https://github.com/josepDe/Project_Aria_STR.

Truong Dang, Tien Thanh Nguyen, John McCall et al.

In recent years, deep learning has rapidly become a method of choice for the segmentation of medical images. Deep Neural Network (DNN) architectures such as UNet have achieved state-of-the-art results on many medical datasets. To further improve the performance in the segmentation task, we develop an ensemble system which combines various deep learning architectures. We propose a two-layer ensemble of deep learning models for the segmentation of medical images. The prediction for each training image pixel made by each model in the first layer is used as the augmented data of the training image for the second layer of the ensemble. The prediction of the second layer is then combined by using a weights-based scheme in which each model contributes differently to the combined result. The weights are found by solving linear regression problems. Experiments conducted on two popular medical datasets namely CAMUS and Kvasir-SEG show that the proposed method achieves better results concerning two performance metrics (Dice Coefficient and Hausdorff distance) compared to some well-known benchmark algorithms.