Abstract: We present a method for the cost estimation of custom hoses from CAD models. They can come in two formats, which are easy to generate: a STL file or the image of a CAD drawing showing several orthogonal projections. The challenges in either cases are, first, to obtain from them a high level 3D description of the shape, and second, to learn a regression function for the prediction of the manufacturing time, based on geometric features of the reconstructed shape. The chosen description is the 3D line along the medial axis of the tube and the diameter of the circular sections along it. In order to extract it from STL files, we have adapted RANSAC, a robust parametric fitting algorithm. As for CAD drawing images, we propose a new technique for 3D reconstruction from data entered on any number of orthogonal projections. The regression function is a Gaussian process, which does not constrain the function to adopt any specific form and is governed by just two parameters. We assess the accuracy of the manufacturing time estimation by k-fold cross validation on 171 STL file models for which the time is provided by an expert. The results show the feasibility of the method, whereby the relative error for 80% of the testing samples is below 15%.

Abstract: A good reliable approach to cardiac triggering is of utmost importance in obtaining accurate quantitative results of atherosclerotic plaque burden from the analysis of IntraVascular UltraSound. Although, in the last years, there has been an increase in research of methods for retrospective gating, there is no general consensus in a validation protocol. Many methods are based on quality assessment of longitudinal cuts appearance and those reporting quantitative numbers do not follow a standard protocol. Such heterogeneity in validation protocols makes faithful comparison across methods a difficult task. We propose a validation protocol based on the variability of the retrieved cardiac phase and explore the capability of several quality measures for quantifying such variability. An ideal detector, suitable for its application in clinical practice, should produce stable phases. That is, it should always sample the same cardiac cycle fraction. In this context, one should measure the variability (variance) of a candidate sampling with respect a ground truth (reference) sampling, since the variance would indicate how spread we are aiming a target. In order to quantify the deviation between the sampling and the ground truth, we have considered two quality scores reported in the literature: signed distance to the closest reference sample and distance to the right of each reference sample. We have also considered the residuals of the regression line of reference against candidate sampling. The performance of the measures has been explored on a set of synthetic samplings covering different cardiac cycle fractions and variabilities. From our simulations, we conclude that the metrics related to distances are sensitive to the shift considered while the residuals are robust against fraction and variabilities as far as one can establish a pair-wise correspondence between candidate and reference. We will further investigate the impact of false positive and negative detections in experimental data.

Abstract: The performances of Sign Language Recognition (SLR) systems have improved considerably in recent years. However, several open challenges still need to be solved to allow SLR to be useful in practice. The research in the field is in its infancy in regards to the robustness of the models to a large diversity of signs and signers, and to fairness of the models to performers from different demographics. This work summarises the ChaLearn LAP Large Scale Signer Independent Isolated SLR Challenge, organised at CVPR 2021 with the goal of overcoming some of the aforementioned challenges. We analyse and discuss the challenge design, top winning solutions and suggestions for future research. The challenge attracted 132 participants in the RGB track and 59 in the RGB+Depth track, receiving more than 1.5K submissions in total. Participants were evaluated using a new large-scale multi-modal Turkish Sign Language (AUTSL) dataset, consisting of 226 sign labels and 36,302 isolated sign video samples performed by 43 different signers. Winning teams achieved more than 96% recognition rate, and their approaches benefited from pose/hand/face estimation, transfer learning, external data, fusion/ensemble of modalities and different strategies to model spatio-temporal information. However, methods still fail to distinguish among very similar signs, in particular those sharing similar hand trajectories.

Abstract: This paper presents results from the second Thermal Image Super-Resolution (TISR) challenge organized in the framework of the Perception Beyond the Visible Spectrum (PBVS) 2021 workshop. For this second edition, the same thermal image dataset considered during the first challenge has been used; only mid-resolution (MR) and high-resolution (HR) sets have been considered. The dataset consists of 951 training images and 50 testing images for each resolution. A set of 20 images for each resolution is kept aside for evaluation. The two evaluation methodologies proposed for the first challenge are also considered in this opportunity. The first evaluation task consists of measuring the PSNR and SSIM between the obtained SR image and the corresponding ground truth (i.e., the HR thermal image downsampled by four). The second evaluation also consists of measuring the PSNR and SSIM, but in this case, considers the x2 SR obtained from the given MR thermal image; this evaluation is performed between the SR image with respect to the semi-registered HR image, which has been acquired with another camera. The results outperformed those from the first challenge, thus showing an improvement in both evaluation metrics.

Abstract: Human drivers produce a vast amount of data which could, in principle, be used to improve autonomous driving systems. Unfortunately, seemingly straightforward approaches for creating end-to-end driving models that map sensor data directly into driving actions are problematic in terms of interpretability, and typically have significant difficulty dealing with spurious correlations. Alternatively, we propose to use this kind of action-based driving data for learning representations. Our experiments show that an affordance-based driving model pre-trained with this approach can leverage a relatively small amount of weakly annotated imagery and outperform pure end-to-end driving models, while being more interpretable. Further, we demonstrate how this strategy outperforms previous methods based on learning inverse dynamics models as well as other methods based on heavy human supervision (ImageNet).