Abstract

Abstract

Abstract
In recent years, the task of person re-identification (Re-ID) has improved considerably with the advances in deep learning methodologies. However, occluded person Re-ID remains a challenging task, as parts of the body of the individual are frequently hidden by various objects, obstacles, or other people, making the identification process more difficult. To address these issues, we introduce a novel data augmentation strategy using artificial occlusions, consisting of random shapes and objects from a small image dataset that was created. We also propose an end-to-end methodology for occluded person Re-ID, which consists of three branches: a global branch, a feature dropping branch, and an occlusion detection branch. Experimental results show that the use of random shape occlusions is superior to random erasing using our architecture. Results on six datasets consisting of three tasks (holistic, partial and occluded person Re-ID) demonstrate that our method performs favourably against state-of-the-art methodologies.

Abstract
With the growing adoption of Deep Learning for imaging tasks in biometrics and healthcare, it becomes increasingly important to ensure privacy when using and sharing images of people. Several works enable privacy-preserving image sharing by anonymizing the images so that the corresponding individuals are no longer recognizable. Most works average images or their embeddings as an anonymization technique, relying on the assumption that the average operation is irreversible. Recently, cold diffusion models, based on the popular denoising diffusion probabilistic models, have succeeded in reversing deterministic transformations on images. In this work, we leverage cold diffusion to decompose superimposed images, empirically demonstrating that it is possible to obtain two or more identically-distributed images given their average. We propose novel sampling strategies for this task and show their efficacy on three datasets. Our findings highlight the risks of averaging images as an anonymization technique and argue for the use of alternative anonymization strategies.

Abstract

Abstract
Breast cancer locoregional treatment can cause significant and long-lasting alterations to a patient's body. As various surgical options may be available to a patient and considering the impact that the aesthetic outcome may have on the patient's self-esteem, it is critical for the patient to be adequately informed of the possible outcomes of each treatment when deciding on the treatment plan. With the purpose of simulating how a patient may look like after treatment, we propose a deep generative model to transfer asymmetries caused by treatment from post-operative breast patients into pre-operative images, taking advantage of the inherent symmetry of breast images. Furthermore, we disentangle asymmetries related with the breast shape from the nipple within the latent space of the network, enabling higher control over the alterations to the breasts. Finally, we show the proposed model's wide applicability in medical imaging, by applying it to generate counterfactual explanations for cardiomegaly and pleural effusion prediction in chest radiographs.