폭간의 기술블로그 HomeArchivesTagsCategories
3D Vision 2
|Machine Learning|

3D Vision

Implicit Function: NeRF

  • NeRF
    • 2020 년에 나온 논문
    • 인풋으로 카메라 포즈를 알고 있는 픽셀들의 레이드를 알고 있어 그걸로 렌더링하는 게 목표
    • 데이터는 신에대한 이미지
    • colmap sfm pipeline을 사용해서 데이터를 만들었다
  • Positional Encoding
    • fourfeat
  • Multiscale Representation
    • mipnerf
  • Learned Initializations
    • learnit
  • Relighting
    • nerv
  • giraffe

Open3D

  • 3D Sensors: Depth Map
  • Structured Light
  • Depth Camera (Prime Sense & Kinect)
  • Lidar Sensor
  • 3D Scene Datasets
    • RGBD dataset
    • ScanNet
    • KITTI
    • CityScape Dataset
    • Semantic KITTI
    • Waymo Open Dataset
    • KITTI 360 Dataset
    • Indoor Lidar Dataset
    • AI Habitat ‒ HM3D
  • 3D Object Datasets
    • ShapeNet
    • ABC Dataset
    • ABO Dataset
    • Google Scanned Object

COLMAP (SfM, MVS)

  • open 3d
  • COLMAP
    • Structure-from-Motion (SfM) and Multi-View Stereo (MVS) pipeline
    • COLMAP SfM
      • • COLMAP’s incremental Structure-from-Motion pipeline.

3D reconstruction

  • Images -> Points: Structure from Motion
  • Points -> More points: Multiple View Stereo
  • Points -> Meshes: Model Fitting
  • Meshes -> Models: Texture Mapping

Human Reconstruction

  • 3D Human Datasets
    • Renderpeople
    • ANIMAR 4

참조

3D Vision