Flank Yang

GaussianObject. GaussianObject: High-Quality 3D Object Reconstruction from Four Views with Gaussian Splatting (SIGGRAPH Asia 2024, TOG)

vehicle_tracking_yolov3. This is the final project of the machine vision course. Given that multi-object tracking (MOT) is currently the research focus, this project will open source to help related research. The basis of this project is: yolo v3, multi-target recognition algorithm, Kalman filter tracking, Hungarian algorithm. (IPIL 2016) This project has basically completed the identification and tracking of vehicle targets in the video, but the robustness is not perfect.

Statistical-Learning-2021. Jupyter Notebook

nerf-fusion. for fusion

compare_cite. for compare citations, which is useful for analysis

pytorch-CycleGAN-and-pix2pix. Image-to-Image Translation in PyTorch

RGBD-Nerf. for RGBD

chensjtu.github.io. simple page

pix2pixHD. Synthesizing and manipulating 2048x1024 images with conditional GANs

kplanes-endo. Python

zoe_pred. for simple zoe depth prediction

iPERCore. Liquid Warping GAN with Attention: A Unified Framework for Human Image Synthesis

AdelaiDepth. This repo contains the projects: 'Virtual Normal', 'DiverseDepth', and '3D Scene Shape'. They aim to solve the monocular depth estimation, 3D scene reconstruction from single image problems.

Poxture. This is offical implement of Poxture: Poxture: Self-supervised Human Posture Imitation using Neural Texture.

learncv_img. used for learning cv

train_fc_layer. 作为一个全连接层的运用，用于估计回归模型

learning_python. this github is made for my learning story

square_img. resize images with pading blank.