Object re-identification and tracking lay the foundation for various computer vision and robotics applications. In this study, we propose a method for personalizing a neural network to enhance and continuously adapt the re-identification of a specific target. Employing an unsupervised continual learning approach in conjunction with an intelligent image pool collection, we can effectively track the target and mitigate the issue of catastrophic forgetting, a challenge prevalent in this research domain. Our primary goal is to provide a robust person re-identification approach to extend the capabilities of recent tracking frameworks employed in robotics, which we have adopted as our baselines for evaluation. Our results demonstrate our approach’s efficacy in successfully re-identifying the target, even when the target drastically changes his clothing appearance and the baseline frameworks struggle. To optimally tune the framework parameters, we conducted an ablation study and substantiated our findings with saliency maps to elucidate the reasons behind the effectiveness of our approach.

In scenarios of Human-Robot Interaction (HRI), it is often assumed that the robot should cooperate with the closest individual or that only one person is present. However, in real-life situations, such as shop floor operations, this assumption may not hold. Thus, it becomes necessary for a robot to recognize a specific target in a crowded environment. To address this problem, we propose a person re-identification module that uses continuous visual adaptation techniques. This module ensures that the robot can seamlessly cooperate with the appropriate individual despite its appearance changes or partial or total occlusions. We used both a laboratory environment and an HRI scenario where the robot followed a person to test our framework. During the test, the targets were asked to change their appearance and disappear from the camera’s field of view to test the module’s ability to handle challenging cases of occlusion and outfit variations. We compared our framework with a state-of-the-art Multi-Object Tracking (MOT) method, and the results showed that our module, shortly named CARPE-ID, accurately tracked each selected target throughout the experiments in all cases except for two cases. In contrast, the MOT had an average of 4 tracking errors for each video.

In state-of-the-art deep single-label classification models, the top- k (k=2,3,4, ...) accuracy is usually significantly higher than the top-1 accuracy. This is more evident in fine-grained datasets, where differences between classes are quite subtle. Exploiting the information provided in the top k predicted classes boosts the final prediction of a model. We propose Guided Zoom, a novel way in which explainability could be used to improve model performance. We do so by making sure the model has "the right reasons" for a prediction. The reason/evidence upon which a deep neural network makes a prediction is defined to be the grounding, in the pixel space, for a specific class conditional probability in the model output. Guided Zoom examines how reasonable the evidence used to make each of the top- k predictions is. Test time evidence is deemed reasonable if it is coherent with evidence used to make similar correct decisions at training time. This leads to better informed predictions. We explore a variety of grounding techniques and study their complementarity for computing evidence. We show that Guided Zoom results in an improvement of a model's classification accuracy and achieves state-of-the-art classification performance on four fine-grained classification datasets. Our code is available at https://github.com/andreazuna89/Guided-Zoom.

This paper addresses nonverbal behavior analysis for the classification of perceived personality traits using novel deep visual activity (VA)-based features extracted only from key-dynamic images. Dynamic images represent short-term VA. Key-dynamic images carry more discriminative information i.e., nonverbal features (NFs) extracted from them contribute to the classification more than NFs extracted from other dynamic images. Dynamic image construction, learning long-term VA with CNN+LSTM, and detecting spatio-temporal saliency are applied to determine key-dynamic images. Once VA-based NFs are extracted, they are encoded using covariance, and resulting representation is used for classification. This method was evaluated on two datasets: small group meetings and vlogs. For the first dataset, proposed method outperforms not only the state-of-the-art VA-based methods but also multi-modal approaches for all personality traits. For extraversion classification, it performs better than i) the most popular key-frames selection algorithm, ii) random and uniform dynamic image selection, and iii) NFs extracted from all dynamic images. Furthermore, the ablation study proves the superiority of proposed method. For the further dataset, it performs as well as the state-of-the-art visual-NFs on average, while showing improved performance for agreeableness classification. Proposed method can be adapted to any application based on nonverbal behavior analysis, thanks to being data-driven.

The concept of compressing deep Convolutional Neural Networks (CNNs) is essential to use limited computation, power, and memory resources on embedded devices. However, existing methods achieve this objective at the cost of a drop in inference accuracy in computer vision tasks. To address such a drawback, we propose a framework that leverages knowledge distillation along with customizable block-wise optimization to learn a lightweight CNN structure while preserving better control over the compression-performance tradeoff. Considering specific resource constraints, e.g., floating-point operations per inference (FLOPs) or model-parameters, our method results in a state of the art network compression while being capable of achieving better inference accuracy. In a comprehensive evaluation, we demonstrate that our method is effective, robust, and consistent with results over a variety of network architectures and datasets, at negligible training overhead. In particular, for the already compact network MobileNet_v2, our method offers up to 2× and 5.2× better model compression in terms of FLOPs and model-parameters, respectively, while getting 1.05% better model performance than the baseline network.

Abstract This paper aims at investigating the action prediction problem from a pure kinematic perspective. Specifically, we address the problem of recognizing future actions, indeed human intentions, underlying a same initial (and apparently unrelated) motor act. This study is inspired by neuroscientific findings asserting that motor acts at the very onset are embedding information about the intention with which are performed, even when different intentions originate from a same class of movements. To demonstrate this claim in computational and empirical terms, we designed an ad hoc experiment and built a new 3D and 2D dataset where, in both training and testing, we analyze a same class of grasping movements underlying different intentions. We investigate how much the intention discriminants generalize across subjects, discovering that each subject tends to affect the prediction by his/her own bias. Inspired by the domain adaptation problem, we propose to interpret each subject as a domain, leading to a novel subject adversarial paradigm. The proposed approach favorably copes with our new problem, boosting the considered baseline features encoding 2D and 3D information and which do not exploit the subject information.

Autism is a behavioral neurological disorder affecting a significant percentage of worldwide population. It especially starts manifesting at very low ages, but it is difficult to early diagnose it since there is not a specific exam or trial that is able to spot it safely. Its detection is in fact mainly dependent from the medical expertise used to assess the patient behavior during direct interviews. This work aims at providing an automatic objective support to the doctor for the assessment of (early) diagnosis of possible autistic subjects by only using video sequences. The underlying idea and rationale come from the psychological and neuroscience studies claiming that the executions of simple motor acts are different between pathological and healthy subjects, and this can be sufficient to discriminate between them. To this end, we devised an experiment in which we recorded, using a standard video camera, patient and healthy children performing the same simple gesture of grasping a bottle. By only processing the video clips depicting the grasping action using a recurrent deep neural network, we are able to discriminate with a good accuracy between the 2 classes of subjects. The designed deep model is also able to provide a sort of attention map in which the zones in the video of major interest are identified in space and time: this “explains” in a certain way which areas the model deems more relevant to the classification purpose, which could also be used by the doctor to make the diagnosis. In the end, this work constitutes a first step towards the development of an automatic computational system devoted to the early diagnosis of autistic subjects, providing the medical expert of a supportive objective method, potentially simple to use in clinical and also more open settings.

A novel method is proposed for generic target tracking by audio measurements from a microphone array. To cope with noisy environments characterized by persistent and high energy interfering sources, a classification map (CM) based on spectral signatures is calculated by means of a machine learning algorithm. Next, the CM is combined with the acoustic map, describing the spatial distribution of sound energy, in order to obtain a cleaned joint map in which contributions from the disturbing sources are removed. A likelihood function is derived from this map and fed to a particle filter yielding the target location estimation on the acoustic image. The method is tested on two real environments, addressing both speaker and vehicle tracking. The comparison with a couple of trackers, relying on the acoustic map only, shows a sharp improvement in performance, paving the way to the application of audio tracking in real challenging environments.

In this paper, we address the new problem of the prediction of human intentions. There is neuro-psychological evidence that actions performed by humans are anticipated by peculiar motor acts which are discriminant of the type of action going to be performed afterwards. In other words, an actual intention can be forecast by looking at the kinematics of the immediately preceding movement. To prove it in a computational and quantitative manner, we devise a new experimental setup where, without using contextual information, we predict human intentions all originating from the same motor act. We posit the problem as a classification task and we introduce a new multi-modal dataset consisting of a set of motion capture marker 3D data and 2D video sequences, where, by only analysing very similar movements in both training and test phases, we are able to predict the underlying intention, i.e., the future, never observed action. We also present an extensive experimental evaluation as a baseline, customizing state-of-the-art techniques for either 3D and 2D data analysis. Realizing that video processing methods lead to inferior performance but show complementary information with respect to 3D data sequences, we developed a 2D+3D fusion analysis where we achieve better classification accuracies, attesting the superiority of the multimodal approach for the context-free prediction of human intentions.

In this paper we aim at increasing the descriptive power of the covariance matrix, limited in capturing linear mutual dependencies between variables only. We present a rigorous and principled mathematical pipeline to recover the kernel trick for computing the covariance matrix, enhancing it to model more complex, non-linear relationships conveyed by the raw data. To this end, we propose Kernelized-COV, which generalizes the original covariance representation without compromising the efficiency of the computation. In the experiments, we validate the proposed framework against many previous approaches in the literature, scoring on par or superior with respect to the state of the art on benchmark datasets for 3D action recognition.