HOSC: A Periodic Activation with Saturation Control for High-Fidelity Implicit Neural Representations
Periodic activations such as sine preserve high-frequency information in implicit neural representations (INRs) through their oscillatory structure, but often suffer from gradient instability and limited control over multi-scale behavior. We introduce the Hyperbolic Oscillator with Saturation Control (HOSC) activation, $\text{HOSC}(x) = \tanh\bigl(β\sin(ω_0 x)\bigr)$, which exposes an explicit parameter $β$ that controls the Lipschitz bound of the activation by $βω_0$. This provides a direct mechanism to tune gradient magnitudes while retaining a periodic carrier. We provide a mathematical analysis and conduct a comprehensive empirical study across images, audio, video, NeRFs, and SDFs using standardized training protocols. Comparative analysis against SIREN, FINER, and related methods shows where HOSC provides substantial benefits and where it achieves competitive parity. Results establish HOSC as a practical periodic activation for INR applications, with domain-specific guidance on hyperparameter selection. For code visit the project page https://hosc-nn.github.io/ .
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
HOSC: A Periodic Activation Function for Preserving Sharp Features in Implicit Neural Representations
Recently proposed methods for implicitly representing signals such as images, scenes, or geometries using coordinate-based neural network architectures often do not leverage the choice of activation functions, or do so o…
Activation Saturation and Floquet Spectrum Collapse in Neural ODEs
We prove that activation saturation imposes a structural dynamical limitation on autonomous Neural ODEs $\dot{h}=f_θ(h)$ with saturating activations ($\tanh$, sigmoid, etc.): if $q$ hidden layers of the MLP $f_θ$ satisfy…
DD-VNB: A Depth-based Dual-Loop Framework for Real-time Visually Navigated Bronchoscopy
Real-time 6 DOF localization of bronchoscopes is crucial for enhancing intervention quality. However, current vision-based technologies struggle to balance between generalization to unseen data and computational speed. I…
Depth EstimationMonocular Depth EstimationMotion EstimationGuidance and Control Networks with Periodic Activation Functions
Inspired by the versatility of sinusoidal representation networks (SIRENs), we present a modified Guidance & Control Networks (G&CNETs) variant using periodic activation functions in the hidden layers. We demonstrate tha…
EchoScene: Indoor Scene Generation via Information Echo over Scene Graph Diffusion
We present EchoScene, an interactive and controllable generative model that generates 3D indoor scenes on scene graphs. EchoScene leverages a dual-branch diffusion model that dynamically adapts to scene graphs. Existing …
3D Object RetrievalDenoisingScene GenerationTexture Synthesis