MEGAN: Memory Enhanced Graph Attention Network for Space-Time Video Super-Resolution

You, Chenyu; Han, Lianyi; Feng, Aosong; Zhao, Ruihan; Tang, Hui; Fan, Wei

doi:10.1109/wacv51458.2022.00400

Cited by 11 publications

(2 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Inaccurate segmentation due to severe metal artifacts or irregular metal shapes may cause some compromises in model training. Fortunately, there are many intelligent segmentation algorithms (You et al 2022a(You et al , 2022c based on deep learning, so to improve the performance of our model, we will consider introducing deep learning-based methods to segment metal traces in future research.…”

Section: Discussionmentioning

confidence: 99%

MARGANVAC: metal artifact reduction method based on generative adversarial network with variable constraints

Li,

Ji,

You

et al. 2023

Phys. Med. Biol.

Self Cite

View full text Add to dashboard Cite

Objective. Metal artifact reduction (MAR) has been a key issue in CT imaging. Recently, MAR methods based on deep learning have achieved promising results. However, when deploying deep learning-based MAR in real-world clinical scenarios, two prominent challenges arise. One limitation is the lack of paired training data in real applications, which limits the practicality of supervised methods. Another limitation is that image-domain methods suitable for more application scenarios are inadequate in performance while end-to-end approaches with better performance are only applicable to fan-beam CT due to large memory consumption. Approach. We propose a novel image-domain MAR method based on the generative adversarial network with variable constraints (MARGANVAC) to improve MAR performance. The proposed variable constraint is a kind of time-varying cost function that can relax the fidelity constraint at the beginning and gradually strengthen the fidelity constraint as the training progresses. To better deploy our image-domain supervised method into practical scenarios, we develop a transfer method to mimic the real metal artifacts by first extracting the real metal traces and then adding them to artifact-free images to generate paired training data. Main results. The effectiveness of the proposed method is validated in simulated fan-beam experiments and real cone-beam experiments. All quantitative and qualitative results demonstrate that the proposed method achieves superior performance compared with the competing methods. Significance. The MARGANVAC model proposed in this paper is an image-domain model that can be conveniently applied to various scenarios such as fan beam and cone beam CT. At the same time, its performance is on par with the cutting-edge dual-domain MAR approaches. In addition, the metal artifact transfer method proposed in this paper can easily generate paired data with real artifact features, which can be better used for model training in real scenarios.

show abstract

Section: Discussionmentioning

confidence: 99%

MARGANVAC: metal artifact reduction method based on generative adversarial network with variable constraints

Li,

Ji,

You

et al. 2023

Phys. Med. Biol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Graph-based Representation for Video Understanding: Previous works have used graphs for action localisation [26,60,63], task completion [25], video super-resolution [61], grounding in instructional videos [23], question answering [55], and action recognition [7,28,37,56]. Hussein et al [26] utilise graphs to analyse human activity from a single video.…”

Section: Related Workmentioning

confidence: 99%

My View is the Best View: Procedure Learning from Egocentric Videos

Bansal

Arora

Jawahar

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Given multiple videos of the same task, procedure learning addresses identifying the key-steps and determining their order to perform the task. For this purpose, existing approaches use the signal generated from a pair of videos. This makes key-steps discovery challenging as the algorithms lack inter-videos perspective. Instead, we propose an unsupervised Graph-based Procedure Learning (GPL) framework. GPL consists of the novel UnityGraph that represents all the videos of a task as a graph to obtain both intra-video and inter-videos context. Further, to obtain similar embeddings for the same key-steps, the embeddings of UnityGraph are updated in an unsupervised manner using the Node2Vec algorithm. Finally, to identify the key-steps, we cluster the embeddings using KMeans. We test GPL on benchmark ProceL, CrossTask, and EgoProceL datasets and achieve an average improvement of 2% on third-person datasets and 3.6% on EgoProceL over the state-of-the-art.

show abstract

TallFormer: Temporal Action Localization with a Long-Memory Transformer

Cheng

Bertasius

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

MEGAN: Memory Enhanced Graph Attention Network for Space-Time Video Super-Resolution

Cited by 11 publications

References 55 publications

MARGANVAC: metal artifact reduction method based on generative adversarial network with variable constraints

MARGANVAC: metal artifact reduction method based on generative adversarial network with variable constraints

My View is the Best View: Procedure Learning from Egocentric Videos

TallFormer: Temporal Action Localization with a Long-Memory Transformer

Contact Info

Product

Resources

About