隐式神经表示(INRs)相关论文汇总

Title: Implicit Neural Representations with Periodic Activation Functions
Date: 2020
Short Title: SIREN
Organization: Stanford University
Journals/Conferences: (NeurIPS)Advances in neural information processing systems
Abstract: 本文利用周期激活函数进行隐式神经表征，称为正弦表征网络，其非常适合表示复杂的自然信号。
Paper
Code

---

Title: Learning Deep Implicit Functions for 3D Shapes with Dynamic Code Clouds
Date: 2022
Short Title: DCC-DIF
Organization: School of Software, BNRist, Tsinghua University
Journals/Conferences: (CVPR)Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
Abstract: 本文将局部编码与可学习的位置向量显式关联起来，这些位置向量是连续的、可动态优化的，提高了隐式网络的表达能力。
Paper
Code

---

Title: MINER: Multiscale Implicit Neural Representation
Date: 2022
Short Title: MINER
Organization: Rice University
Journals/Conferences: (ECCV)European Conference on Computer Vision
Abstract: 本文的多尺度隐式神经表示(MINER)通过拉普拉斯金字塔进行内部表示，它提供了信号的稀疏多尺度分解，能够捕捉跨尺度信号的正交部分。
Paper
Code

---

Title: Acorn: Adaptive Coordinate Networks for Neural Scene Representation
Date: 2022
Short Title: ACORN
Organization: Stanford University
Journals/Conferences: (ECCV)European Conference on Computer Vision
Abstract: 本文引入了一种新的混合隐-显网络结构和训练策略，该策略在训练和推理过程中根据感兴趣信号的局部复杂性自适应分配资源。
Paper
Code

---

Title: Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
Date: 2020
Short Title: FPE
Organization: University of California
Journals/Conferences: (NeurIPS)Advances in neural information processing systems
Abstract: 本文证明，通过一个简单的傅里叶特征映射来传递输入点，可以使多层感知器(MLP)学习低维问题域中的高频函数。
Paper
Code

---

Title: Transformers as Meta-Learners for Implicit Neural Representations
Date: 2022
Short Title: Trans-INR
Organization: UC San Diego
Journals/Conferences: (ECCV)European Conference on Computer Vision
Abstract: 本文使用Transformers作为INR的超网络，在这里它可以直接构建整个INR权重集，Transformers被专门用作集到集映射。
Paper
Code