In contrast to the widely studied problem of recognizing an action given a complete sequence, action anticipation aims to identify the action from only partially available videos. As such, it is therefore key to the success of computer vision applications requiring to react as early as possible, such as autonomous navigation. In this paper, we propose a new action anticipation method that achieves high prediction accuracy even in the presence of a very small percentage of a video sequence. To this end, we develop a multi-stage LSTM architecture that leverages context-aware and action-aware features, and introduce a novel loss function that encourages the model to predict the correct class as early as possible. Our experiments on standard benchmark datasets evidence the benefits of our approach; We outperform the state-of-the-art action anticipation methods for early prediction by a relative increase in accuracy of 22.0% on JHMDB-21, 14.0% on UT-Interaction and 49.9% on UCF-101.
With a questionnaire addressed to general dental practitioners in Sweden, the Swedish Council on Technology Assessment in Health Care launched a project group in 1999 to systematically review and evaluate the existing literature on various caries preventive methods. The aim of this article was to report findings concerning the caries preventive effect of fluoride toothpastes in various age groups, with special emphasis on fluoride concentration and supervised versus non-supervised brushing. A systematic search in electronic databases for articles published between 1966 and April 2003 was conducted with the inclusion criteria of a randomized or controlled clinical trial, at least 2 years follow-up and caries increment in the permanent (deltaDMFS/T) or primary (deltadmfs/t) dentition as endpoint. Out of 905 articles originally identified, 54 met the inclusion criteria. These studies were assessed independently by at least two reviewers and scored A-C according to predetermined criteria for methodology and performance. The measure of effect was the prevented fraction (PF), expressed as percent. The results revealed strong evidence (level 1) (i) for the caries preventive effect of daily use of fluoride toothpaste compared to placebo in the young permanent dentition (PF 24.9%), (ii) that toothpastes with 1,500 ppm of fluoride had a superior preventive effect compared with standard dentifrices with 1,000 ppm F in the young permanent dentition (PF 9.7%), and (iii) that higher caries reductions were recorded in studies with supervised toothbrushing compared with non-supervised (PF 23.3%). However, incomplete evidence (level 4) was found regarding the effect of fluoride toothpaste in the primary dentition. In conclusion, this review reinforced the importance of daily toothbrushing with fluoridated toothpastes for preventing dental caries, although long-term studies in age groups other than children and adolescents are still lacking.
Pixel-level annotations are expensive and time consuming to obtain. Hence, weak supervision using only image tags could have a significant impact in semantic segmentation. Recently, CNN-based methods have proposed to fine-tune pre-trained networks using image tags. Without additional information, this leads to poor localization accuracy. This problem, however, was alleviated by making use of objectness priors to generate foreground/background masks. Unfortunately these priors either require training pixel-level annotations/bounding boxes, or still yield inaccurate object boundaries. Here, we propose a novel method to extract markedly more accurate masks from the pre-trained network itself, forgoing external objectness modules. This is accomplished using the activations of the higher-level convolutional layers, smoothed by a dense CRF. We demonstrate that our method, based on these masks and a weakly-supervised loss, outperforms the state-of-the-art tag-based weakly-supervised semantic segmentation techniques. Furthermore, we introduce a new form of inexpensive weak supervision yielding an additional accuracy boost.
The objectives of this study were to evaluate systematically the evidence of the caries-preventive effect of fissure sealing of occlusal tooth surfaces and to examine factors potentially modifying the effect. The search strategies included electronic databases, reference lists of articles, and selected textbooks. Inclusion criteria were randomized or quasi-randomized clinical trials or controlled clinical trials comparing fissure sealing with no treatment or another preventive treatment in children up to 14 years of age at the start; the outcome measure was caries increment; the diagnostic criteria had been described; and the follow-up time was at least 2 years. Inclusion decisions were taken and grading of the studies was done independently by two of the authors. The main measure of effect was relative risk reduction. Thirteen studies using resin-based or glass ionomer sealant materials were included in the final analysis. The results showed that most studies were performed during the 1970s and a single application had been utilized. The relative caries risk reduction pooled estimate of resin-based sealants on permanent 1st molars was 33% (relative risk = 0.67; CI = 0.55-0.83). The effect depended on retention of the sealant. In conclusion, the review suggests limited evidence that fissure sealing of 1st permanent molars with resin-based materials has a caries-preventive effect. The evidence is incomplete for permanent 2nd molars, premolars and primary molars and for glass ionomer cements. Overall, there remains a need for further trials of high quality, particularly in child populations with a low and a high caries risk, respectively.
Gaussian mixture alignment is a family of approaches that are frequently used for robustly solving the point-set registration problem. However, since they use local optimisation, they are susceptible to local minima and can only guarantee local optimality. Consequently, their accuracy is strongly dependent on the quality of the initialisation. This paper presents the first globally-optimal solution to the 3D rigid Gaussian mixture alignment problem under the L 2 distance between mixtures. The algorithm, named GOGMA, employs a branch-and-bound approach to search the space of 3D rigid motions SE(3), guaranteeing global optimality regardless of the initialisation. The geometry of SE(3) was used to find novel upper and lower bounds for the objective function and local optimisation was integrated into the scheme to accelerate convergence without voiding the optimality guarantee. The evaluation empirically supported the optimality proof and showed that the method performed much more robustly on two challenging datasets than an existing globally-optimal registration solution.
We demonstrate that many detection methods are designed to identify only a sufficently accurate bounding box, rather than the best available one. To address this issue we propose a simple and fast modification to the existing methods called Fitness NMS. This method is tested with the DeNet model and obtains a significantly improved MAP at greater localization accuracies without a loss in evaluation rate, and can be used in conjunction with Soft NMS for additional improvements. Next we derive a novel bounding box regression loss based on a set of IoU upper bounds that better matches the goal of IoU maximization while still providing good convergence properties. Following these novelties we investigate RoI clustering schemes for improving evaluation rates for the DeNet wide model variants and provide an analysis of localization performance at various input image dimensions. We obtain a MAP of 33.6%@79Hz and 41.8%@5Hz for MSCOCO and a Titan X (Maxwell).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.