Bus Violence: An Open Benchmark for Video Violence Detection on Public Transport.

The automatic detection of violent actions in public places through video analysis is difficult because the employed Artificial Intelligence-based techniques often suffer from generalization problems. Indeed, these algorithms hinge on large quantities of annotated data and usually experience a drastic drop in performance when used in scenarios never seen during the supervised learning phase. In this paper, we introduce and publicly release the Bus Violence benchmark, the first large-scale collection of video clips for violence detection on public transport, where some actors simulated violent actions inside a moving bus in changing conditions, such as the background or light. Moreover, we conduct a performance analysis of several state-of-the-art video violence detectors pre-trained with general violence detection databases on this newly established use case. The achieved moderate performances reveal the difficulties in generalizing from these popular methods, indicating the need to have this new collection of labeled data, beneficial for specializing them in this new scenario.

The Face Deepfake Detection Challenge.

Multimedia data manipulation and forgery has never been easier than today, thanks to the power of Artificial Intelligence (AI). AI-generated fake content, commonly called Deepfakes, have been raising new issues and concerns, but also new challenges for the research community. The Deepfake detection task has become widely addressed, but unfortunately, approaches in the literature suffer from generalization issues. In this paper, the Face Deepfake Detection and Reconstruction Challenge is described. Two different tasks were proposed to the participants: (i) creating a Deepfake detector capable of working in an "in the wild" scenario; (ii) creating a method capable of reconstructing original images from Deepfakes. Real images from CelebA and FFHQ and Deepfake images created by StarGAN, StarGAN-v2, StyleGAN, StyleGAN2, AttGAN and GDWCT were collected for the competition. The winning teams were chosen with respect to the highest classification accuracy value (Task I) and "minimum average distance to Manhattan" (Task II). Deep Learning algorithms, particularly those based on the EfficientNet architecture, achieved the best results in Task I. No winners were proclaimed for Task II. A detailed discussion of teams' proposed methods with corresponding ranking is presented in this paper.

Virtual to Real Adaptation of Pedestrian Detectors.

Pedestrian detection through Computer Vision is a building block for a multitude of applications. Recently, there has been an increasing interest in convolutional neural network-based architectures to execute such a task. One of these supervised networks' critical goals is to generalize the knowledge learned during the training phase to new scenarios with different characteristics. A suitably labeled dataset is essential to achieve this purpose. The main problem is that manually annotating a dataset usually requires a lot of human effort, and it is costly. To this end, we introduce ViPeD (Virtual Pedestrian Dataset), a new synthetically generated set of images collected with the highly photo-realistic graphical engine of the video game GTA V (Grand Theft Auto V), where annotations are automatically acquired. However, when training solely on the synthetic dataset, the model experiences a Synthetic2Real domain shift leading to a performance drop when applied to real-world images. To mitigate this gap, we propose two different domain adaptation techniques suitable for the pedestrian detection task, but possibly applicable to general object detection. Experiments show that the network trained with ViPeD can generalize over unseen real-world scenarios better than the detector trained over real-world data, exploiting the variety of our synthetic dataset. Furthermore, we demonstrate that with our domain adaptation techniques, we can reduce the Synthetic2Real domain shift, making the two domains closer and obtaining a performance improvement when testing the network over the real-world images.

Effects of UV-B irradiation on a marine microecosystem.

Purpose of this work was to study the effect of UV irradiation on a microecosystem consisting of several interacting species. The system chosen was of a hypersaline type, where all the species present live at high salt concentration; it comprises different bacteria; a producer, the photosynthetic green alga Dunaliella salina; and a consumer, the ciliated protozoan Fabrea salina, which form a complete food chain. We were able to establish the initial conditions that give rise to a self-sustaining microecosystem, stable for at least 3 weeks. We then determined the effect of UV irradiation on this microecosystem under laboratory-controlled conditions, in particular by measuring the critical UV exposure for the two main components of the microecosystem (algae and protozoa) under UV-B irradiances comparable to those of solar irradiation. In our experiments, we varied irradiance, total dose and spectral composition of the actinic light. The critical doses at irradiances of the order of 56 kJ/m(2) (typical average daily irradiance in a sunny summer day in Pisa), measured for each main component of the microecosystem (algae and ciliates), turned out to be around 70 kJ/m(2) for ciliates and 50 kJ/m(2) for D. salina. By exposing microecosystems to daily UV-B irradiances of the order of 8 kJ/m(2) (typical average daily irradiance in a sunny winter day in Pisa), we found no effect at total doses of the order of the critical doses at high irradiances, showing that the reciprocity law does not hold. We have also measured a preliminary spectral-sensitive curve of the UV effects, which shows an exponential decay with wavelength.