Supannee Tanathong's Projects
Unsupervised anomaly detection with generative model, keras implementation
A fork repo of https://github.com/BayesRays/BayesRays for experimenting with NeRFs uncertainty
Caffe: a fast open framework for deep learning. This repository was forked from https://github.com/BVLC/caffe.
This repository was a fork of https://github.com/TimoSaemann/caffe-segnet-cudnn5 which is based on https://github.com/alexgkendall/caffe-segnet. In this fork, a new layer: CustomBlobMultinomialLogisticLoss is added (which works with BVLC/caffe as well).
Implementation of multinomial logistic loss that accepts a probability blob of size NxCxHxW and a label blob of size Nx1xHxC.
Self-driving car nanodegree course project
A helper class to check/set on a defined DWORD struct in which images on an image sequence a (ground) point exists.
Forked repository -- COLMAP - Structure-from-Motion and Multi-View Stereo
This repository stores C/C++ codes for various data structure and algorithms which I have been practicing on to further develop my skills.
Repo for the Deep Learning Nanodegree Foundations program.
A collection of online tools for a quick work on browser for research/development
A handy tool to extract camera-pose in the form of projection matrix given Reconstruction.json, the output of OpenSfM (https://github.com/mapillary/OpenSfM)
This photogrammetric project is to detect fiducial marks on image in order to determine the interior orientation parameters of the camera.
List of image stitching approaches including papers and sourcecode
This project is to visualize LiDAR data in 2.5D (virtual top down view) with the ability to zoom and pan.
This was forked from https://github.com/matterport/Mask_RCNN/. Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow
matplotlib: plotting with Python
A list of tools for multiple camera calibration
Open Source Computer Vision Library
This repository is forked from openMVG/openMVG. -- open Multiple View Geometry library. Basis for 3D computer vision and Structure from Motion.
A utility to extract camera poses [R|t] from OpenMVG's SfM results
This repository is forked from [mapillary/OpenSfM](https://github.com/mapillary/OpenSfM) Open source Structure from Motion pipeline
Sourcecode to produce undistorted image in which all source image pixels are retained. Two methods are implemented (1) using cv::initUndistortRectifyMap and cv::getOptimalNewCameraMatrix, (2) using the reverse distortion technique with cv::remap to produce the undistorted image.
The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
List of SfM based on its category
A simple deep neural network implemented in C++,based with OpenCV Mat matrix class
Forked repository for GPU-accelerated real-time stixel computation (http://www.cvc.uab.es/people/dhernandez)
Supplementary material for "HighwayView: Seamless, tile-based orthomosaics from millions of street level images"
Jupyter Notebook for study topics required for CNN