Search

Estimating depth from RGB images is a long-standing ill-posed problem, which has been explored for decades by the computer vision, graphics, and machine learning communities. Among the existing techniques, stereo matching remains one of the most widely used in the literature due to its strong connection to the human binocular system. Traditionally, stereo-based depth estimation has been addressed through matching hand-crafted features across multiple images. Despite the extensive amount of research, these traditional techniques still suffer in the presence of highly textured areas, large uniform regions, and occlusions. Motivated by their growing success in solving various 2D and 3D vision problems, deep learning for stereo-based depth estimation has attracted a growing interest from the community, with more than 150 papers published in this area between 2014 and 2019. This new generation of methods has demonstrated a significant leap in performance, enabling applications such as autonomous driving and augmented reality. In this paper, we provide a comprehensive survey of this new and continuously growing field of research, summarize the most commonly used pipelines, and discuss their benefits and limitations. In retrospect of what has been achieved so far, we also conjecture what the future may hold for deep learning-based stereo for depth estimation research. [ABSTRACT FROM AUTHOR]

Bycatch remains one of the most important issues in the world's fisheries so its estimation and reporting have been highlighted in many international, regional and jurisdictional guidelines and policies. This paper describes a simple methodology to estimate jurisdictional discards, using Australia's first national bycatch report as a case study. The methodology involves: (a) identifying annual landings for all fisheries and methods; (b) deriving retained:discard ratios for each; (c) where ratios are lacking, using substitute ratios from similar fisheries; (d) applying the ratios from (b) and (c) to the data from (a) to obtain totals; and (e) scoring the quality of the discard information using the US Tier Classification System weighted by estimated discard levels. The results for Australia revealed that, during the last decade, commercial fisheries annually discarded 42.5% of what was caught (87,983t). 70% came from just eight fisheries/methods with 30% coming from the other 299. The Queensland East Coast Prawn Trawl fishery contributed 28.5% of the national total. The quality of discard information was reasonable across most jurisdictions, with a national score of 59.1%. The best quality data came from the Commonwealth due to its observer and (more recent) Electronic Monitoring programmes. Those data also showed that fishers' logbook information under‐estimated levels of discards (determined from observer data) by 89.7%. This paper provides: (a) the means to develop benchmarks in bycatch management and estimation against which jurisdictions can be compared and performances tracked; and (b) for Australia, priority areas for management intervention to reduce discarding and improve its monitoring. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel approach for 3-D rain field reconstruction using satellite signals. It uses the estimated signal-to-noise ratio (SNR) at the ground receivers for low-earth orbit (LEO) satellites and, thus, indirectly estimates the path-integrated rain attenuation of the microwave communication links. A least-squares algorithm is employed to perform the 3-D tomographic reconstruction of the rain field. The proposed system model consists of an LEO satellite with a realistic overpass trajectory and multiple ground receivers with SNR estimators. Two synthetic rain events near the Great Barrier Reef in Australia are used to test the reconstruction outcome. Simulation results suggest that the reconstructed rain field has close agreement with the synthetic rain field. [ABSTRACT FROM AUTHOR]

Performance evaluation of multi-target tracking algorithms is of great practical importance in the design, parameter optimization and comparison of tracking systems. The goal of performance evaluation is to measure the distance between two sets of tracks: the ground truth tracks and the set of estimated tracks. This paper proposes a mathematically rigorous metric for this purpose. The basis of the proposed distance measure is the recently formulated consistent metric for performance evaluation of multi-target filters, referred to as the OSPA metric. Multi-target filters sequentially estimate the number of targets and their position in the state space. The OSPA metric is therefore defined on the space of finite sets of vectors. The distinction between filtering and tracking is that tracking algorithms output tracks and a track represents a labeled temporal sequence of state estimates, associated with the same target. The metric proposed in this paper is therefore defined on the space of finite sets of tracks and incorporates the labeling error. Numerical examples demonstrate that the proposed metric behaves in a manner consistent with our expectations. [ABSTRACT FROM PUBLISHER]

Learning occurs in all stages of a fishery, in the entire fishing career of a fisher, and even within a single fishing trip. It is an important aspect of fleet dynamics; failure to take it into proper account biases results of population dynamics studies and results of stock assessments. In this paper, I introduce and slightly generalise three learning models and derive a new one. I use them to analyse data on the short-term learning behaviour of all fishers as a group in the fishery of the western king prawn Penaeus latisulcatus in the Gulf St Vincent, Australia, to demonstrate their utilities. The damped periodic change inferred in fishing efficiency of all fishers as a group with time (shot number) over a fishing day in the P. latisulcatus fishery had a period of 9.44±1.06 shots and could have resulted from periodic ‘forgetting’ and forcing by good catch rates. The new learning model was used to test the hypothesis that fishers in the P. latisulcatus fishery learn more from their failures in catch rate than from their successes, or more from their successes than from their failures, or as much from their successes as from their failures. Although very limited, the data suggest that fishers as a group learn 69.03±311.52% more from their successes in good catch rates than from their failures. [Copyright &y& Elsevier]