Category: Softwares & Codes

Links to code and softwares by me.

Hands on TensorRT on NvidiaTX2

Resources: Official Base Page: https://developer.nvidia.com/tensorrtOfficial User Guide: https://docs.nvidia.com/deeplearning/sdk/tensorrt-api/python_api/index.htmlWebminar: Introduction to TensorRT. http://on-demand.gputechconf.com/gtcdc/2017/video/DC7172 My Codes involving TensorRTMinimalist demo for Keras model --> .pb --> .uff [GIT]C++ API for TensorRT5, works on TX2 [GIT] Introduction You are most probably familiar with deep learning frameworks like Tensorflow, Pytorch, mxnet etc. These frameworks are general purpose tools geared towards … Continue reading Hands on TensorRT on NvidiaTX2 →

Optimal Triangulation for Tuning Keypoint Co-ordinates

Given a set of correct keypoint matches and a fundamental matrix, to optimize the coordinates of these key points such that they satisfy the epipolar constraint. A point (x,y) on the left image (pose: [I|0]) and (x',y') on the right image (pose: [R|t]). These points are undistorted and in normalized image coordinates. Having known the pose … Continue reading Optimal Triangulation for Tuning Keypoint Co-ordinates →

Image Keypoint Descriptors and Matching

[GitHub] Extracting keypoints from images, usually, corner points etc is usually the first step for geometric methods in computer vision. A typical workflow is: keypoints are extracted from images (SIFT, SURF, ORB etc.). At these keypoints descriptors are extracted (SURF, ORB etc). Usually a 32D vector at each keypoint. The nearest neighbor search is performed to … Continue reading Image Keypoint Descriptors and Matching →

HowTo – Pose Graph Bundle Adjustment

SLAM (Simultaneous Localization and Mapping) is one of the important practical areas in computer vision / robotics / image based modelling community. A SLAM system typically consists of a) odometry estimator (relative pose estimator), b) Bundle adjustment module, c) sensor fusion module (for visual-inertial system), d) mapping module. While there are several excellent resources, refer … Continue reading HowTo – Pose Graph Bundle Adjustment →

Parallel Bulk Image Resizing – Python

Usually multi-threaded programming can be typically. One typically will need a queue to maintain the inputs. From this thread-safe queue the threads pick an input, act on it and save the result. This can usually take up most of the afternoon depending on the job. This recipe is adapted from [link]. It introduces a way … Continue reading Parallel Bulk Image Resizing – Python →

Vision Controlled Quadcopter Flight

Just completed (8th Dec, 2015) the Aerial Robotics course in HKUST (ELEC6910P) by Prof. SHEN Shaojie (my PhD supervisor). Most of the course credit were on the completion of the projects. Course TAs were Su Kunyue and Yang Zhengfei. My project partners were Sun Lin and Sun Ting The projects eventually connected together to make a vision … Continue reading Vision Controlled Quadcopter Flight →

Direct Edge Alignment (D-EA)

Abstract There has been a paradigm shifting trend towards feature-less methods due to their elegant formulation, accuracy and ever increasing computational power. In this work, we present a direct edge alignment approach for 6-DOF tracking. We argue that photo-consistency based methods are plagued by a much smaller convergence basin and are extremely sensitive to noise, changing illumination … Continue reading Direct Edge Alignment (D-EA) →

Color Consistency from Multiple-View Images

Abstract In this paper we address the problem of multiview color consistency. We propose to use a graph model of 3d positions obtained using matched dense feature points. We define an energy functional on this model which captures relationship of the colors across views while also imposing a smoothness cost to obtain optimal colors for the 3d … Continue reading Color Consistency from Multiple-View Images →