Ziyi Li

Ziyi Li

Auto-generated profile, based on publicly available information

Ziyi Li

Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 July 2023 in IEEE Transactions on Learning Technologies
Verified
Journal Article
Closed access
QMQiusha Min
Qiusha Min
ZZZhongwei Zhou
Zhongwei Zhou
ZLZiyi Li
Ziyi Li
MWMei Wu
Mei Wu
Instructional videos are often a key component of online learning, and their quality significantly influences online learning outcomes and student satisfaction. However, instructional video evaluation is time-consuming. To solve this problem, this study developed an automatic evaluation method for instructional videos. This method first establishes a metric to evaluate instructional videos based on two aspects: video features and watching experience. An automatic scoring method for each indicator was developed based on video and clickstream data. Finally, all the scores were input into the evaluation model to obtain the evaluation result. Our experimental results showed that 85% of the evaluation results using our proposed model are consistent with manual quality evaluation. Therefore, our method can perform automatic evaluation of instructional videos while achieving acceptable accuracy, which is helpful in reducing the workload associated with manual evaluation and improving the quality of online teaching.

7 October 2022 in arXiv
Verified
Preprint
Open access
PDF file
QZQinye Zhou
Qinye Zhou
ZLZiyi Li
Ziyi Li
WXWeidi Xie
Weidi Xie
XZXiaoyun Zhang
Xiaoyun Zhang
YZYa Zhang
Ya Zhang
YWYanfeng Wang
Yanfeng Wang
Existing models on super-resolution often specialized for one scale, fundamentally limiting their use in practical scenarios. In this paper, we aim to develop a general plugin that can be inserted into existing super-resolution models, conveniently augmenting their ability towards Arbitrary Resolution Image Scaling, thus termed ARIS. We make the following contributions: (i) we propose a transformer-based plugin module, which uses spatial coordinates as query, iteratively attend the low-resolution image feature through cross-attention, and output visual feature for the queried spatial location, resembling an implicit representation for images; (ii) we introduce a novel self-supervised training scheme, that exploits consistency constraints to effectively augment the model's ability for upsampling images towards unseen scales, i.e. ground-truth high-resolution images are not available; (iii) without loss of generality, we inject the proposed ARIS plugin module into several existing models, namely, IPT, SwinIR, and HAT, showing that the resulting models can not only maintain their original performance on fixed scale factor but also extrapolate to unseen scales, substantially outperforming existing any-scale super-resolution models on standard benchmarks, e.g. Urban100, DIV2K, etc.