Efficient long-short temporal modeling is key for enhancing the performance of action recognition task. In this paper, we propose a new two-stream action recognition network, termed as MENet, consisting of a Motion Enhancement (ME) module and a Video-level Aggregation (VLA) module to achieve long-short temporal modeling. Specifically, motion representations have been proved effective in capturing short-term and high-frequency action. However, current motion representations are calculated from adjacent frames, which may have poor interpretation and bring useless information (noisy or blank). Thus, for short-term motions, we design an efficient ME module to enhance the short-term motions by mingling the motion saliency among neighboring segments. As for long-term aggregations, VLA is adopted at the top of the appearance branch to integrate the long-term dependencies across all segments. The two components of MENet are complementary in temporal modeling. Extensive experiments are conducted on UCF101 and HMDB51 benchmarks, which verify the effectiveness and efficiency of our proposed MENet.

Multi-modal Ads Video Understanding Challenge is the first grand challenge aiming to comprehensively understand ads videos. Our challenge includes two tasks: video structuring and multi-label classification. Video structuring asks the participants to accurately predict both the scene boundaries and the multi-label categories of each scene based on a fine-grained and ads-related category hierarchy. This task will advance the foundation of comprehensive ads video understanding, which has a significant impact on many applications in ads, such as video recommendation and user behavior analysis. This paper presents an overview of the video structuring task in our grand challenge, including the background of ads videos, an elaborate description of this task, our proposed dataset, the evaluation protocol, and our baseline model. By ablating the key components of our baseline, we would like to reveal the main challenges of this task and provide useful guidance for future research of this area.

A better understanding of wildfire spreading processes is expected both for long-term wildfire management planning and effective suppression strategies so that wildfires-caused losses in natural resources, endangered species, human lives and properties can be reduced. The cellular automaton (CA) method is useful and powerful for modeling natural and artificial systems in which the constituent parts are locally and tightly interacted. The temporal and spatial characteristics of wildfires make the CA method a preferred choice to model the temporal and spatial variability. In this study, hexagonal tessellation space was used to develop a CA model and then the Rothermers fire spreading model (1972) was integrated to the CA model to simulate wildfire-spreading processes spatially and temporally. The hexagon-based wildfire spreading CA model was verified with the experimental data and also tested using the data obtained in the southern part of Gansu province (China), and the comparison between the simulated results and observed data suggests that this CA-based model can be used to provide useful information for managing wildfires over heterogeneous landscapes.