paper-with-me

홈 › Papers

Revisiting the Scale Loss Function and Gaussian-Shape Convolution for Infrared Small Target Detection

2026-04-11 · Hao Li, Man Fung Zhuo arxiv

Infrared small target detection still faces two persistent challenges: training instability from non-monotonic scale loss functions, and inadequate spatial attention due to generic convolution kernels that ignore the physical imaging characteristics of small targets. In this paper, we revisit both aspects. For the loss side, we propose a \emph{diff-based scale loss} that weights predictions according to the signed area difference between the predicted mask and the ground truth, yielding strictly monotonic gradients and stable convergence. We further analyze a family of four scale loss variants to understand how their geometric properties affect detection behavior. For the spatial side, we introduce \emph{Gaussian-shaped convolution} with a learnable scale parameter to match the center-concentrated intensity profile of infrared small targets, and augment it with a \emph{rotated pinwheel mask} that adaptively aligns the kernel with target orientation via a straight-through estimator. Extensive experiments on IRSTD-1k, NUDT-SIRST, and SIRST-UAVB demonstrate consistent improvements in $mIoU$, $P_d$, and $F_a$ over state-of-the-art methods. We release our anonymous code and pretrained models.

📄 PDF Abstract BibTeX arXiv:2604.09991

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

PG-3DGS: Optimizing 3D Gaussian Splatting to Satisfy Physics Objectives

2026-05-11 · Zachary Lee, Maxwell Jacobson, Yexiang Xue arxiv

Recent advances in Gaussian Splatting have enabled fast, high-fidelity 3D scene generation, yet these methods remain purely visual and lack an understanding of how shapes behave in the physical world. We introduce Physic…

Scene Generation

3D-GMNet: Single-View 3D Shape Recovery as A Gaussian Mixture

2019-12-10 · Kohei Yamashita, Shohei Nobuhara, Ko Nishino

In this paper, we introduce 3D-GMNet, a deep neural network for 3D object shape reconstruction from a single image. As the name suggests, 3D-GMNet recovers 3D shape as a Gaussian mixture. In contrast to voxels, point clo…

3D ReconstructionDensity EstimationPose Estimation

Revisiting Map Relations for Unsupervised Non-Rigid Shape Matching

2023-10-17 · Dongliang Cao, Paul Roetzer, Florian Bernard

We propose a novel unsupervised learning approach for non-rigid 3D shape matching. Our approach improves upon recent state-of-the art deep functional map methods and can be applied to a broad range of different challengi…

Scale-Parameter Selection in Gaussian Kolmogorov-Arnold Networks

2026-04-23 · Amir Noorizadegan, Sifan Wang arxiv

Kolmogorov--Arnold Networks (KANs) have recently attracted attention as edge-based neural architectures in which learnable univariate functions replace conventional fixed activation functions. A key source of flexibility…

Revisiting "Qualitatively Characterizing Neural Network Optimization Problems"

2020-12-12 · NeurIPS Workshop DL-IG 2020 12 · Jonathan Frankle

We revisit and extend the experiments of Goodfellow et al. (2014), who showed that - for then state-of-the-art networks - "the objective function has a simple, approximately convex shape" along the linear path between in…