arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.15355cs.CV
收藏

注意力机制优于线性方法:快速隐式神经表示生成

Attention Beats Linear for Fast Implicit Neural Representation Generation

College of Computer Science, Zhejiang University(浙江大学计算机学院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室)

Shuyi Zhang, Ke Liu, Jingjun Gu, Xiaoxu Cai, Zhihua Wang, Jiajun Bu, Haishuai Wang

AI总结:

本文提出基于注意力局部化的隐式神经表示(ANR),结合局部注意力层与全局MLP,利用超网络生成紧凑表示,提升重建精度与效率,在CelebA上PSNR提升至47.25dB。

摘要:

隐式神经表示(INR)作为一种数据表示方法日益受到关注,成为创新生成模型的前提。与基于梯度的方法在推理时效率较低不同,采用超网络为执行INR函数的多层感知机(MLP)生成参数,已成为一种有前景且高效的替代方案。然而,MLP作为全局连续函数,难以建模高度不连续的信号,导致训练阶段收敛缓慢且重建性能不准确。此外,MLP需要大量表示参数,这意味着数据表示的效率低下。本文提出了一种新颖的基于注意力的局部化INR(ANR),由局部注意力层(LAL)和全局MLP组成,将坐标特征与数据特征整合并转换为有意义的输出。随后,我们设计了一个实例表示框架,该框架提供类似Transformer的超网络,将数据实例表示为紧凑的表示向量。借助实例特定的表示向量和实例无关的ANR参数,目标信号被很好地重建为连续函数。我们进一步通过变分坐标解决获取超分辨率推理结果时的混叠伪影。在四个数据集上的广泛实验展示了ANR方法的显著效果,例如在CelebA数据集上将PSNR值从37.95dB提升至47.25dB。代码已发布在https://github.com/Roninton/ANR。

英文摘要:

Implicit Neural Representation (INR) has gained increasing popularity as a data representation method, serving as a prerequisite for innovative generation models. Unlike gradient-based methods, which exhibit lower efficiency in inference, the adoption of hyper-network for generating parameters in Multi-Layer Perceptrons (MLP), responsible for executing INR functions, has surfaced as a promising and efficient alternative. However, as a global continuous function, MLP is challenging in modeling highly discontinuous signals, resulting in slow convergence during the training phase and inaccurate reconstruction performance. Moreover, MLP requires massive representation parameters, which implies inefficiencies in data representation. In this paper, we propose a novel Attention-based Localized INR (ANR) composed of a localized attention layer (LAL) and a global MLP that integrates coordinate features with data features and converts them to meaningful outputs. Subsequently, we design an instance representation framework that delivers a transformer-like hyper-network to represent data instances as a compact representation vector. With instance-specific representation vector and instance-agnostic ANR parameters, the target signals are well reconstructed as a continuous function. We further address aliasing artifacts with variational coordinates when obtaining the super-resolution inference results. Extensive experimentation across four datasets showcases the notable efficacy of our ANR method, e.g. enhancing the PSNR value from 37.95dB to 47.25dB on the CelebA dataset. Code is released at https://github.com/Roninton/ANR.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑