paper-with-me

홈 › Papers

Time-series Initialization and Conditioning for Video-agnostic Stabilization of Video Super-Resolution using Recurrent Networks

2024-03-23 · Hiroshi Mori, Norimichi Ukita

A Recurrent Neural Network (RNN) for Video Super Resolution (VSR) is generally trained with randomly clipped and cropped short videos extracted from original training videos due to various challenges in learning RNNs. However, since this RNN is optimized to super-resolve short videos, VSR of long videos is degraded due to the domain gap. Our preliminary experiments reveal that such degradation changes depending on the video properties, such as the video length and dynamics. To avoid this degradation, this paper proposes the training strategy of RNN for VSR that can work efficiently and stably independently of the video length and dynamics. The proposed training strategy stabilizes VSR by training a VSR network with various RNN hidden states changed depending on the video properties. Since computing such a variety of hidden states is time-consuming, this computational cost is reduced by reusing the hidden states for efficient training. In addition, training stability is further improved with frame-number conditioning. Our experimental results demonstrate that the proposed method performed better than base methods in videos with various lengths and dynamics.

📄 PDF Abstract BibTeX arXiv:2403.15832

Code (0)

등록된 구현이 없습니다.

Tasks

Super-ResolutionTime SeriesVideo Super-Resolution

Methods 이 논문이 사용한 방법론

BASE 설명 없음

Similar Papers 제목 키워드 기반

Non-autoregressive Conditional Diffusion Models for Time Series Prediction

2023-06-08 · Lifeng Shen, James Kwok

Recently, denoising diffusion models have led to significant breakthroughs in the generation of images, audio and text. However, it is still an open question on how to adapt their strong modeling ability to model time se…

DenoisingOpen-Ended Question AnsweringTime SeriesTime Series Prediction

Does Semantic Noise Initialization Transfer from Images to Videos? A Paired Diagnostic Study

2026-03-03 · Yixiao Jing, Chaoyu Zhang, Zixuan Zhong, Peizhou Huang arxiv

Semantic noise initialization has been reported to improve robustness and controllability in image diffusion models. Whether these gains transfer to text-to-video (T2V) generation remains unclear, since temporal coupling…

Semantically-Guided Inference for Conditional Diffusion Models: Enhancing Covariate Consistency in Time Series Forecasting

2025-08-03 · Rui Ding, Hanyang Meng, Zeyang Zhang, Jielong Yang arxiv

Diffusion models have demonstrated strong performance in time series forecasting, yet often suffer from semantic misalignment between generated trajectories and conditioning covariates, especially under complex or multim…

Time Series Forecasting

Multimodal Meta-Learning for Time Series Regression

2021-08-05 · Sebastian Pineda Arango, Felix Heinrich, Kiran Madhusudhanan, Lars Schmidt-Thieme

Recent work has shown the efficiency of deep learning models such as Fully Convolutional Networks (FCN) or Recurrent Neural Networks (RNN) to deal with Time Series Regression (TSR) problems. These models sometimes need a…

Meta-LearningregressionTime SeriesTime Series Analysis+1

Test-Time Self-Adaptive Conditioning for Stable Audio-Driven Talking-Head Generation

2026-05-25 · Zhicheng Zhang, Lei Wang, Yu Zhang, Yongsheng Gao arxiv

Audio-driven talking-head generation has achieved remarkable progress with recent models such as AniTalker, FLOAT, and Sonic. Despite their success, most existing approaches rely on a single static reference image to con…

Video Generation