使用薄板样条(TPS)的空间变换网络(STN)的 PyTorch 实现
thin-plate-splines 开源项目推荐(20 个)
thin-plate-splines 话题在 GitHub 上没有填写简介,这里按 Star 数汇总最相关的热门仓库。
热门项目 全部 20
GitHub 共 20 个仓库,已全部展示DALF is a joint image keypoint detector and descriptor for handling non-rigid deformations. It also works great under large rotations.
PyTorch implementation of the descriptor DEAL presented at NeurIPS 2021 "Extracting Deformation-Aware Local Features by Learning to Deform".
Geometric Morphometrics in Python
Interactive thin-plate spline interpolation
A research-grade tool for comprehensive spatial and descriptive analysis of soils.
"Parallax-Tolerant Image Stitching Based on Robust Elastic Warping" paper implementation
Source code for R package chebpol on CRAN
C# / .NET wrapper for thin plate spline interpolation based on Peter Stroia-Williams C++ implementation
Python code to swap faces from images and videos using classical and deep learning approach.
A python implementation of an end-to-end pipeline for swapping faces from images and videos using classical and deep learning approach.
CMSC733 - Computer Processing Pictorial Information: End-to-end pipeline to swap faces in videos and images using warping methods Delaunay Triangulation and Thin Plate Spline
Python implementation of an end-to-end pipeline for swapping faces from images and videos using classical and deep learning approach
该项目在 GitHub 上没有填写描述
DeepKriging on the Global Data
Functional interpolation to create a more meaningful interpolation than bilinear or bicubic methods.
Implementation of face swapping using classical and deep learning technique.
Implementing a Python-based project using computer vision to analyse a wooden beer bottle cap display (shaped like Germany), aiming to correlate bottle caps with their geographic references by minimizing spatial distances between map holes and physical locations.
Computationally efficient Bayesian variable selection method to identify linear and nonlinear effects in regression models where a predictor (or feature) can have a zero effect, a non-zero linear effect or non-zero non-linear effect on the outcome variables.
A hybrid restoration framework for ancient documents that combines Deep Learning (U-Net) for structural mask detection with classical geometric transformations (TPS, Hough Transform) to rectify distortions while ensuring 100% historical integrity and verifiability.
关于这个话题
- 清单取自 GitHub 话题页并按 Star 数排序,每次展示前 20 个项目;原始页面见 https://github.com/topics/thin-plate-splines。
- 该话题在 GitHub 上共有 20 个仓库, 本站收录的是其中最热门的 20 个,不是全量列表。
- Star / Fork 每日自动刷新一次,实时数据以 GitHub 为准。
- 当前 20 个项目累计 1136 Star,覆盖 6 种主要语言,数据更新于 2026-11-10。
- 中文介绍来自 GitHub 官方话题说明(github/explore)的机器翻译,可点「查看原文」对照;译文缺失时会先展示原文。