Eduardo A. B. da Silva

Rui Lourenço, Lucas Thomaz, Eduardo A. B. Silva et al.

Light field cameras and multi-camera arrays have emerged as promising solutions for accurately estimating depth by passively capturing light information. This is possible because the 3D information of a scene is embedded in the 4D light field geometry. Commonly, depth estimation methods extract this information relying on gradient information, heuristic-based optimisation models, or learning-based approaches. This paper focuses mainly on explicitly understanding and exploiting 4D geometrical cues for light field depth estimation. Thus, a novel method is proposed, based on a non-learning-based optimisation approach for depth estimation that explicitly considers surface normal accuracy and occlusion regions by utilising a fully explainable 4D geometric model of the light field. The 4D model performs depth/disparity estimation by determining the orientations and analysing the intersections of key 2D planes in 4D space, which are the images of 3D-space points in the 4D light field. Experimental results show that the proposed method outperforms both learning-based and non-learning-based state-of-the-art methods in terms of surface normal angle accuracy, achieving a Median Angle Error on planar surfaces, on average, 26.3$\%$ lower than the state-of-the-art, and still being competitive with state-of-the-art methods in terms of MSE ${\times}$ 100 and Badpix 0.07.

Wesley L. Passos, Gabriel M. Araujo, Amaro A. de Lima et al.

Every year, the Aedes aegypti mosquito infects millions of people with diseases such as dengue, zika, chikungunya, and urban yellow fever. The main form to combat these diseases is to avoid mosquito reproduction by searching for and eliminating the potential mosquito breeding grounds. In this work, we introduce a comprehensive dataset of aerial videos, acquired with an unmanned aerial vehicle, containing possible mosquito breeding sites. All frames of the video dataset were manually annotated with bounding boxes identifying all objects of interest. This dataset was employed to develop an automatic detection system of such objects based on deep convolutional networks. We propose the exploitation of the temporal information contained in the videos by the incorporation, in the object detection pipeline, of a spatio-temporal consistency module that can register the detected objects, minimizing most false-positive and false-negative occurrences. Also, we experimentally show that using videos is more beneficial than only composing a mosaic using the frames. Using the ResNet-50-FPN as a backbone, we achieve F$_1$-scores of 0.65 and 0.77 on the object-level detection of `tires' and `water tanks', respectively, illustrating the system capabilities to properly locate potential mosquito breeding objects.