CrimeNet: Neural Structured Learning using Vision Transformer for violence detection.

The state of the art in violence detection in videos has improved in recent years thanks to deep learning models, but it is still below 90% of average precision in the most complex datasets, which may pose a problem of frequent false alarms in video surveillance environments and may cause security guards to disable the artificial intelligence system. In this study, we propose a new neural network based on Vision Transformer (ViT) and Neural Structured Learning (NSL) with adversarial training. This network, called CrimeNet, outperforms previous works by a large margin and reduces practically to zero the false positives. Our tests on the four most challenging violence-related datasets (binary and multi-class) show the effectiveness of CrimeNet, improving the state of the art from 9.4 to 22.17 percentage points in ROC AUC depending on the dataset. In addition, we present a generalisation study on our model by training and testing it on different datasets. The obtained results show that CrimeNet improves over competing methods with a gain of between 12.39 and 25.22 percentage points, showing remarkable robustness.

Self-Supervised Learning Across Domains.

Human adaptability relies crucially on learning and merging knowledge from both supervised and unsupervised tasks: the parents point out few important concepts, but then the children fill in the gaps on their own. This is particularly effective, because supervised learning can never be exhaustive and thus learning autonomously allows to discover invariances and regularities that help to generalize. In this paper we propose to apply a similar approach to the problem of object recognition across domains: our model learns the semantic labels in a supervised fashion, and broadens its understanding of the data by learning from self-supervised signals on the same images. This secondary task helps the network to focus on object shapes, learning concepts like spatial orientation and part correlation, while acting as a regularizer for the classification task over multiple visual domains. Extensive experiments confirm our intuition and show that our multi-task method, combining supervised and self-supervised knowledge, provides competitive results with respect to more complex domain generalization and adaptation solutions. It also proves its potential in the novel and challenging predictive and partial domain adaptation scenarios.

Learning Categories From Few Examples With Multi Model Knowledge Transfer.

Learning a visual object category from few samples is a compelling and challenging problem. In several real-world applications collecting many annotated data is costly and not always possible. However, a small training set does not allow to cover the high intraclass variability typical of visual objects. In this condition, machine learning methods provide very few guarantees. This paper presents a discriminative model adaptation algorithm able to proficiently learn a target object with few examples by relying on other previously learned source categories. The proposed method autonomously chooses from where and how much to transfer information by solving a convex optimization problem which ensures to have the minimal leave-one-out error on the available training set. We analyze several properties of the described approach and perform an extensive experimental comparison with other existing transfer solutions, consistently showing the value of our algorithm.

Kernel methods for melanoma recognition.

Skin cancer is a spreading disease in the western world. Early detection and treatment are crucial for improving the patient survival rate. In this paper we present two algorithms for computer assisted diagnosis of melanomas. The first is the support vector machines algorithm, a state-of-the-art large margin classifier, which has shown remarkable performances on object recognition and categorization problems. The second method, spin glass-Markov random fields, combines results of statistical physics of spin glasses with Markov random fields. We compared the two approaches using color histograms as features. We benchmarked our methods with another algorithm presented in the literature, which uses a sophisticated segmentation technique and a set of features especially designed for melanoma recognition. To our knowledge, this algorithm represents the state of the art on skin lesions classification. We show with extensive experiments that the support vector machines approach outperforms the existing method and, on two classes out of three, it achieves performances comparable to those obtained by expert clinicians.