Correspondences between 3D keypoints generated by matching local descriptors are a key step in 3D computer vision and graphic applications. Learned descriptors are rapidly evolving and outperforming the classical handcrafted approaches in the field. Yet, to learn effective representations they require supervision through labeled data, which are cumbersome and time-consuming to obtain. Unsupervised alternatives exist, but they lag in performance. Moreover, invariance to viewpoint changes is attained either by relying on data augmentation, which is prone to degrading upon generalization on unseen datasets, or by learning from handcrafted representations of the input which are already rotation invariant but whose effectiveness at training time may significantly affect the learned descriptor. We show how learning an equivariant 3D local descriptor instead of an invariant one can overcome both issues. LEAD (Local EquivAriant Descriptor) combines Spherical CNNs to learn an equivariant representation together with plane-folding decoders to learn without supervision. Through extensive experiments on standard surface registration datasets, we show how our proposal outperforms existing unsupervised methods by a large margin and achieves competitive results against the supervised approaches, especially in the practically very relevant scenario of transfer learning.

People who have motor disabilities associated with some difficulty in speech demand alternative means to interact with other people and the environment in which they are inserted. Augmentative and Alternative Communication (AAC) refers to all communication forms that can complement or replace speech. The practice of AAC mediated by computational applications represents a very attractive alternative. For people with motor and speech difficulties, gestural interaction can be a way to make interaction with AAC systems feasible. In this thesis, a methodology called MyPGI (Methodology to yield Personalized Gestural Interaction) was developed and validated to guide the design of AAC systems for people with motor and speech difficulties in order to promote greater autonomy for them in handling computer systems and for monitoring by their caregivers. The methodology uses Computer Vision and Machine Learning techniques to enable non-invasive and personalized gestural interaction using low-cost devices. MyPGI was applied and evaluated in real case studies with people with disabilities, informing the design of an interactive system named PGCA (Personal Gesture Communication Assistant) to allow the creation and use of personalized gestural languages for people with severe motor and speech difficulties. This thesis, defended at the Federal University of Parana's Graduate Program in Informatics, shows conceptual, methodological, and technical contributions, with publications in high-level vehicles, registered software, and social technology available for free and open use.

We propose a novel, non-invasive method for identification of human fingerprints obtained from 3D images of the dermis and epidermis. Using images obtained with optical coherence tomography, we compute the Gaussian (K) and mean (H) curvature values of the dermatoglyphics of dermal and epidermal images from volunteers' fingers, which are then converted into curvature-type and KH maps. Next, the regions of maps located around the minutiae are matched based on their normalized cross correlation. To the best of our knowledge, this is the first work to explore the use of KH maps of the finger dermatoglyphics and 3D images of the dermis to support fingerprint identification. The reliability of 3D dermal fingerprints is further explored by comparing unrolled 2D fingerprints extracted from 3D dermal point clouds against a 2D fingerprint test database obtained using a 2D commercial scanner and software. Finally, an experiment is performed to illustrate the robustness of the dermal fingerprint to mild alterations of the epidermis.

Motivation is a key element in the learning process, and using games for this goal is an undeniably attractive option. However, individuals with motor disabilities are still finding themselves excluded from many game-based school approaches. This paper presents a solution developed for people with motor and speech impairments that includes a game-based approach to contribute in this context. Based on a methodology in which gestures and their meanings are created and configured by users and their caregivers, we developed a Computer Vision system named PGCA that employs machine learning techniques to create personalized gestural interaction as an Assistive Technology resource for communication purposes. This personalized gestural interaction enabled disabled people to interact with the PGCA system and play a game using their remaining functional movements. Results from experiments carried out with the target audience indicate that the structure used in the developed system can be expanded to create new games, with different purposes, including to educational context.