paper-with-me

홈 › Papers

B-Spline Texture Coefficients Estimator for Screen Content Image Super-Resolution

2023-01-01 · CVPR 2023 1 · Byeonghyun Pak, Jaewon Lee, Kyong Hwan Jin

Screen content images (SCIs) include many informative components, e.g., texts and graphics. Such content creates sharp edges or homogeneous areas, making a pixel distribution of SCI different from the natural image. Therefore, we need to properly handle the edges and textures to minimize information distortion of the contents when a display device's resolution differs from SCIs. To achieve this goal, we propose an implicit neural representation using B-splines for screen content image super-resolution (SCI SR) with arbitrary scales. Our method extracts scaling, translating, and smoothing parameters of B-splines. The followed multi-layer perceptron (MLP) uses the estimated B-splines to recover high-resolution SCI. Our network outperforms both a transformer-based reconstruction and an implicit Fourier representation method in almost upscaling factor, thanks to the positive constraint and compact support of the B-spline basis. Moreover, our SR results are recognized as correct text letters with the highest confidence by a pre-trained scene text recognition network. Source code is available at https://github.com/ByeongHyunPak/btc.

📄 PDF Abstract BibTeX

Code (1)

byeonghyunpak/btc 공식 구현 pytorch

Tasks

Image Super-ResolutionScene Text RecognitionSuper-Resolution

Similar Papers 제목 키워드 기반

Unwarping Screen Content Images via Structure-texture Enhancement Network and Transformation Self-estimation

2025-04-21 · Zhenzhen Xiao, Heng Liu, Bingwen Hu

While existing implicit neural network-based image unwarping methods perform well on natural images, they struggle to handle screen content images (SCIs), which often contain large geometric distortions, text, symbols, a…

Estimation of Latent Group Structures in Time-Varying Panel Data Models

2025-03-29 · Paul Haimerl, Stephan Smeekes, Ines Wilms

We introduce a panel data model where coefficients vary both over time and the cross-section. Slope coefficients change smoothly over time and follow a latent group structure, being homogeneous within but heterogeneous a…

Deep neural networks for smooth approximation of physics with higher order and continuity B-spline base functions

2022-01-03 · Kamil Doległo, Anna Paszyńska, Maciej Paszyński, Leszek Demkowicz

This paper deals with the following important research question. Traditionally, the neural network employs non-linear activation functions concatenated with linear operators to approximate a given physical phenomenon. Th…

UKAN: Unbound Kolmogorov-Arnold Network Accompanied with Accelerated Library

2024-08-20 · Alireza Moradzadeh, Lukasz Wawrzyniak, Miles Macklin, Saee G. Paliwal

In this work, we present a GPU-accelerated library for the underlying components of Kolmogorov-Arnold Networks (KANs), along with an algorithm to eliminate bounded grids in KANs. The GPU-accelerated library reduces the c…

BenchmarkingComputational EfficiencyGPUKolmogorov-Arnold Networks

Trainable Spline Representations for Physics-Informed Learning

2026-07-17 · Giovanni Canali, Nicola Demo, Gianluigi Rozza arxiv

This work introduces Physics-Informed Splines (PI-Splines), a structured spline-based architecture for physics-informed learning. Instead of representing the solution of a differential equation with a neural network, PI-…