【英文写作日知录 第1期】句式汇总 Sun, 05 July 2020

论文写作学习
为了更好地撰写论文，将从一些优秀的论文、博客和文章中摘出句子进行句式汇总。以便学习和在论文中使用。
👉 ✈更多精彩请移步主页

1.句式汇总

• the problem of xxx has been addressed by xxx

• To accomplish this we train a machine learning model in an end-to-end manner to do

• The xxx architecture was chosen since it has been successful for a number of xxx and xxx tasks.

• The model consists of three parts: a, b, c

• Passing each frame through a ResNet-based encoder yields a sequence of embeddings.

• At this point we calculate XXX by doing sth, returning a matrix that is adj for xxx to do

• we then use xxx

• One advantage of using the xx as xxx layer in xx is that

• We overcome this challenge by introducing a process for xxx that

• To address this, we perform

• it is still too simple for deep learning methods

• The challenge in this is two-fold. First, it requires one to

• they still need to

2.例句

• We present an approach for estimating the period with which an action is repeated in a video.

• The crux of the approach lies in constraining the period prediction module to use temporal self-similarity as an intermediate representation bottleneck that allows generalization to unseen repetitions in videos in the wild.

• We train this model, called RepNet, with a synthetic dataset that is generated from a large unlabeled video collection by sampling short clips of varying lengths and repeating them with different periods and counts.

• This combination of synthetic data and a powerful yet constrained model, allows us to predict periods in a class-agnostic fashion.

• Our model substantially exceeds the state of the art performance on existing periodicity (PERTUBE) and repetition counting (QUVA) benchmarks.

---

ref:
1.https://ai.googleblog.com/2020/06/repnet-counting-repetitions-in-videos.html
2.https://openaccess.thecvf.com/content_CVPR_2020/papers/Dwibedi_Counting_Out_Time_Class_Agnostic_Video_Repetition_Counting_in_the_CVPR_2020_paper.pdf