paper-with-me

홈 › Papers

STAR: A Structure-Aware Lightweight Transformer for Real-Time Image Enhancement

2021-01-01 · ICCV 2021 10 · Zhaoyang Zhang, Yitong Jiang, Jun Jiang, Xiaogang Wang, Ping Luo, Jinwei Gu

Image and video enhancement such as color constancy, low light enhancement, and tone mapping on smartphones is challenging because high-quality images should be achieved efficiently with a limited resource budget. Unlike prior works that either used very deep CNNs or large Transformer models, we propose a \underline s eman\underline t ic-\underline a wa\underline r e lightweight Transformer, termed STAR, for real-time image enhancement. STAR is formulated to capture long-range dependencies between image patches, which naturally and implicitly captures the semantic relationships of different regions in an image. STAR is a general architecture that can be easily adapted to different image enhancement tasks. Extensive experiments show that STAR can effectively boost the quality and efficiency of many tasks such as illumination enhancement, auto white balance, and photo retouching, which are indispensable components for image processing on smartphones. For example, STAR reduces model complexity and improves image quality compared to the recent state-of-the-art [??] on the MIT-Adobe FiveK dataset [??] (i.e., 1.8dB PSNR improvements with 25% parameters and 13% float operations.)

📄 PDF Abstract BibTeX

Code (1)

zzyfd/STAR-pytorch pytorch

Tasks

Color ConstancyImage EnhancementPhoto RetouchingTone MappingVideo Enhancement

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Position-Wise Feed-Forward Layer 설명 없음
Adam 설명 없음
Residual Connection 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

Star-Transformer

2019-02-25 · NAACL 2019 6 · Qipeng Guo, Xipeng Qiu, PengFei Liu, Yunfan Shao 외

Although Transformer has achieved great successes on many NLP tasks, its heavy structure with fully-connected attention connections leads to dependencies on large training data. In this paper, we present Star-Transformer…

Named Entity Recognition (NER)Natural Language InferenceSentiment AnalysisText Classification

SAMformer: Unlocking the Potential of Transformers in Time Series Forecasting with Sharpness-Aware Minimization and Channel-Wise Attention

2024-02-15 · Romain Ilbert, Ambroise Odonnat, Vasilii Feofanov, Aladin Virmaux 외

Transformer-based architectures achieved breakthrough performance in natural language processing and computer vision, yet they remain inferior to simpler linear baselines in multivariate long-term forecasting. To better …

Time SeriesTime Series Forecasting

面向微博文本的融合字词信息的轻量级命名实体识别(Lightweight Named Entity Recognition for Weibo Based on Word and Character)

2020-10-01 · CCL 2020 10 · Chun Chen, Mingyang Li, Fang Kong

中文社交媒体命名实体识别由于其领域特殊性,一直广受关注。非正式且无结构的微博文本存在以下两个问题:一是词语边界模糊;二是语料规模有限。针对问题一,本文将同维度的字词进行融合,获得丰富的文本序列表征;针对问题二,提出了基于Star-Transformer框架的命名实体识别模型,借助星型拓扑结构更好地捕获动态特征;同时利用高速网络优化Star-Transformer中的信息桥接,提升模型的鲁棒性。本文提出的轻量级命名实体识别模型取得了目前W…

named-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)

EdgeFlex-Transformer: Transformer Inference for Edge Devices

2025-12-17 · Shoaib Mohammad, Guanqun Song, Ting Zhu arxiv

Deploying large-scale transformer models on edge devices presents significant challenges due to strict constraints on memory, compute, and latency. In this work, we propose a lightweight yet effective multi-stage optimiz…

Structure-Aware Piano Accompaniment via Style Planning and Dataset-Aligned Pattern Retrieval

2026-02-16 · Wanyu Zang, Yang Yu, Meng Yu arxiv

We introduce a structure-aware approach for symbolic piano accompaniment that decouples high-level planning from note-level realization. A lightweight transformer predicts an interpretable, per-measure style plan conditi…