This paper introduces a video dataset of spatiotemporally localized Atomic Visual Actions (AVA). The AVA dataset densely annotates 80 atomic visual actions in 430 15-minute video clips, where actions are localized in space and time, resulting in 1.58M action labels with multiple labels per person occurring frequently. The key characteristics of our dataset are: (1) the definition of atomic visual actions, rather than composite actions;(2) precise spatio-temporal annotations with possibly multiple annotations for each person; (3) exhaustive annotation of these atomic actions over 15-minute video clips; (4) people temporally linked across consecutive segments; and (5) using movies to gather a varied set of action representations. This departs from existing datasets for spatio-temporal action recognition, which typically provide sparse annotations for composite actions in short video clips.AVA, with its realistic scene and action complexity, exposes the intrinsic difficulty of action recognition. To benchmark this, we present a novel approach for action localization that builds upon the current state-of-the-art methods, and demonstrates better performance on JHMDB and UCF101-24 categories. While setting a new state of the art on existing datasets, the overall results on AVA are low at 15.6% mAP, underscoring the need for developing new approaches for video understanding.

Anaerobic mono- and co-digestion of kitchen waste (KW), corn stover (CS), and chicken manure (CM) under mesophilic (37 °C) conditions were conducted in batch mode with the aim of investigating the biomethane potential (BMP), biodegradability, methane production performance, and stability of the process. An initial volatile solid concentration of 3 g VS L–1 with a substrate-to-inoculum (S/I) ratio of 0.5 was first tested, and two S/I ratios of 1.5 and 3.0 were evaluated subsequently. The modified Gompertz equation was used to assist in the interpretation of the conclusions. The results showed that BMP and specific methane yields were 725 and 683 mL g–1 VSadded for KW, 470 and 214 mL g–1 VSadded for CS, and 617 and 291 mL g–1 VSadded for CM, respectively. Therefore, KW had the highest biodegradability of 94% as compared with CS (45%) or CM (47%). For KW mono- and co-digestion with CS, CM, or their mixture, methane production performance was better at an S/I ratio of 1.5 than that of 3.0. For CS, CM, and their mixture, S/I ratios of both 1.5 and 3.0 were suitable. A synergistic effect was found in the co-digestion process, which was mainly attributed to a proper carbon-to-nitrogen ratio and the reduced total volatile fatty acids-to-total alkalinity ratio, thus providing better buffering capacity and supporting more microorganisms for efficient digestion.

No abstract

No abstract

We present a neural network model -based on Convolutional Neural Networks, Recurrent Neural Networks and a novel attention mechanism -which achieves 84.2% accuracy on the challenging French Street Name Signs (FSNS) dataset, significantly outperforming the previous state of the art (Smith'16), which achieved 72.46%. Furthermore, our new method is much simpler and more general than the previous approach. To demonstrate the generality of our model, we show that it also performs well on an even more challenging dataset derived from Google Street View, in which the goal is to extract business names from store fronts. Finally, we study the speed/accuracy tradeoff that results from using CNN feature extractors of different depths. Surprisingly, we find that deeper is not always better (in terms of accuracy, as well as speed). Our resulting model is simple, accurate and fast, allowing it to be used at scale on a variety of challenging real-world text extraction problems.