论文撰写-部分论文摘要写法分析

1，ECCV’2016-Temporal Segment Networks for Action

Abstract—Deep convolutional networks have achieved great success for image recognition. However, for action recognition in videos,their advantage over traditional methods is not so evident.
【研究背景，动机】先扬后抑。当前深度卷积网络效果很好了，但是在视频的动作识别领域优势还不明显。

【We present a general and flexible video-level framework for learning action models in videos. This method, called temporal segment network (TSN), 】【aims to model long-range temporal structures with a new segment-based sampling and aggregation module. 】【This unique design enables our TSN to efficiently learn action models by using the whole action videos. The learned models could be easily adapted for action recognition in both trimmed and untrimmed videos with simple average pooling and multi-scale temporal window integration, respectively. We also study a series of good practices for the instantiation of temporal segment network framework given limited training samples.】
【研究方法】我们做了什么，想要得到什么的效果，我们的方法优点。

Our approach obtains the state-the-of-art performance on four challenging action recognition benchmarks: HMDB51 (71.0%), UCF101 (94.9%), THUMOS14 (80.1%), and ActivityNet v1.2 (89.6%). Using the proposed RGB difference for motion models, our method can still achieve competitive accuracy on UCF101 (91.0%) while running at 340 FPS. Furthermore, based on the temporal segment networks, we won the video classification track at the ActivityNet challenge 2016 among 24 teams, which demonstrates the effectiveness of temporal segment network and the proposed good practices.
【研究结果】我们的方法在基准数据集上有什么的准确率，表明了什么。
Index Terms—Action Recognition; Temporal Segment Networks; Temporal Modeling; Good Practices; ConvNets 【关键词】行为识别领域，时间片段网络【我们提出来的】，时间建模，很好实例，卷积网络【我们做的】

2，[2017 non-local] Non-local Neural Networks

Both convolutional and recurrent operations are building blocks that process one localneighborhood at a time.【研究背景，现状】

In this paper, we present non-local operations as a generic family of building blocks for capturing long-range dependencies.
【研究方法1】【present…for…】

Inspired by the classical non-local means method [4] in computer vision, our non-local operation computes the response at a position as a weighted sum of the features at all positions.
【研究方法2】方法的inspiration,way (how to do it)

This building block can be plugged into many computer vision architectures. On the task of video classification, even without any bells and whistles, our nonlocal models can compete or outperform current competition winners on both Kinetics and Charades datasets.
【研究结果】 此处虽然没有具体的准确率，但是从优点，outperform等词看出实验结果很好。

In static image recognition, our non-local models improve object detection/segmentation and pose estimation on the COCO suite of tasks. Code will be made available.

3,[2019] TSM: Temporal Shift Module for Efficient Video Understanding

The explosive growth in video streaming gives rise to challenges on performing video understanding at high accuracy and low computation cost. 【研究背景-视频理解面临挑战】

Conventional 2D CNNs are computationally cheap but cannot capture temporal relationships; 3D CNN based methods can achieve good performance but are computationally intensive, making it expensive to deploy. 【是研究动机也是inspiration】

In this paper, we propose a generic and effective Temporal Shift Module (TSM) that enjoys both high efficiency and high performance. Specifically, it can achieve the performance of 3D CNN but maintain 2D CNN’s complexity. 【研究的主题，优点【即能实现的目标】】

TSM shifts part of the channels along the temporal dimension; thus facilitate information exchanged among neighboring frames. It can be inserted into 2D CNNs to achieve temporal modeling at zero computation and zero parameters. 【研究方法与优点】

We also extended TSM to online setting, which enables real-time low-latency online video recognition and video object detection. 【研究方法与优点】

TSM is accurate and efficient: it ranks the first place on the Something-Something leaderboard upon publication; on Jetson Nano and Galaxy Note8, it achieves a low latency of 13ms and 35ms for online video recognition.【研究结果】不仅仅要写研究结果，还要突出你的优势，伴随有这样的词汇。outperform,rank the first place

The code is available at: https://github.com/mit-han-lab/temporal-shift-module.

4，Deep social force network for anomaly event

Abstract: Anomaly event detection is vital in surveillance video analysis. However, how to learn the discriminative motion in the crowd scene is still not tackled.【研究背景，动机】这个套路就是做什么很重要，然而哪个地方还存在挑战。接下来说我们的研究正解决这个挑战就显得合理。

In this work, we propose a deep social force network by exploiting both social force extracting and deep motion coding.

Given a grid of particles with velocity provided by the optical flow, we investigate the interaction force in the crowd scene and embed a social force module in a deep network.【研究方法1】

We further design a Deep Motion Convolution with a 3D (DMC-3D) module. The DMC-3D not only eliminates the noise motion in the crowd scene with a spatial encoder-decoder but also learn the 3D feature with a spatio-temporal encoder. 【研究方法与优点2】

We model our deep social force coding with multiple features, in which each feature can describe specific anomaly motion. 【研究方法，优点3】

The experiments on the benchmark dataset of UCF-Crime demonstrate that our method can predict the temporal localization of anomaly events and outperform the state-of-the-art【实验结果】在实验室里，在某某数据集跑的实验证明它可以实现什么。