angeladai / 3dmv
Licence: other
[ECCV'18] 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation
Stars: ✭ 161
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to 3dmv
Cnncomplete
[CVPR'17] Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis
Stars: ✭ 117 (-27.33%)
Mutual labels: computer-graphics
Psraytracing
A (modern) C++ implementation of the first two books of the Peter Shirley Ray Tracing mini-books (https://raytracing.github.io). Features a clean project structure, perf. improvements (compared to the original code), multi-core rendering, and more.
Stars: ✭ 129 (-19.88%)
Mutual labels: computer-graphics
Flame pytorch
This is a implementation of the 3D FLAME model in PyTorch
Stars: ✭ 153 (-4.97%)
Mutual labels: computer-graphics
Tiny Renderer
A tiny sotfware 3D renderer in 100 lines of Python
Stars: ✭ 120 (-25.47%)
Mutual labels: computer-graphics
Cyclegan
Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.
Stars: ✭ 10,933 (+6690.68%)
Mutual labels: computer-graphics
Top 10 Computer Vision Papers 2020
A list of the top 10 computer vision papers in 2020 with video demos, articles, code and paper reference.
Stars: ✭ 132 (-18.01%)
Mutual labels: computer-graphics
L1stabilizer
🎥 Video stabilization using L1-norm optimal camera paths.
Stars: ✭ 111 (-31.06%)
Mutual labels: computer-graphics
Awesome Computer Graphics
🔺😎 A curated list of awesome stuff to learn computer graphics
Stars: ✭ 154 (-4.35%)
Mutual labels: computer-graphics
Cubbyflow
Voxel-based fluid simulation engine for computer games
Stars: ✭ 151 (-6.21%)
Mutual labels: computer-graphics
Tinyrenderer
A brief computer graphics / rendering course
Stars: ✭ 11,776 (+7214.29%)
Mutual labels: computer-graphics
Optcuts
OptCuts, a new parameterization algorithm, jointly optimizes arbitrary embeddings for seam quality and distortion. OptCuts requires no parameter tuning; automatically generating mappings that minimize seam-lengths while satisfying user-requested distortion bounds.
Stars: ✭ 145 (-9.94%)
Mutual labels: computer-graphics
Gpu Motunui
GPU-Motunui is a path tracer that renders Disney Animation's Moana Island scene.
Stars: ✭ 120 (-25.47%)
Mutual labels: computer-graphics
Computer Vision Video Lectures
A curated list of free, high-quality, university-level courses with video lectures related to the field of Computer Vision.
Stars: ✭ 154 (-4.35%)
Mutual labels: computer-graphics
Sgnn
[CVPR'20] SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans
Stars: ✭ 112 (-30.43%)
Mutual labels: computer-graphics
Vulkantutorial
Tutorial for the Vulkan graphics and compute API
Stars: ✭ 1,962 (+1118.63%)
Mutual labels: computer-graphics
Giraffe
Distributed Monte Carlo Path Tracer 分布蒙特卡洛路径追踪器
Stars: ✭ 155 (-3.73%)
Mutual labels: computer-graphics
Mvstudio
An integrated SfM (Structure from Motion) and MVS (Multi-View Stereo) solution.
Stars: ✭ 154 (-4.35%)
Mutual labels: computer-graphics
Lowpolify
Create low-poly art from any image 🌟🌟
Stars: ✭ 149 (-7.45%)
Mutual labels: computer-graphics
3DMV
3DMV jointly combines RGB color and geometric information to perform 3D semantic segmentation of RGB-D scans. This work is based on our ECCV'18 paper, 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation.
Code
Installation:
Training is implemented with PyTorch. This code was developed under PyTorch 0.2 and recently upgraded to PyTorch 0.4.
Training:
- See
python train.py --help
for all train options. Example train call:
python train.py --gpu 0 --train_data_list [path to list of train files] --data_path_2d [path to 2d image data] --class_weight_file [path to txt file of train histogram] --num_nearest_images 5 --model2d_path [path to pretrained 2d model]
- Trained models: models.zip
Testing
- See
python test.py --help
for all test options. Example test call:
python test.py --gpu 0 --scene_list [path to list of test scenes] --model_path [path to trained model.pth] --data_path_2d [path to 2d image data] --data_path_3d [path to test scene data] --num_nearest_images 5 --model2d_orig_path [path to pretrained 2d model]
Data:
This data has been precomputed from the ScanNet (v2) dataset.
- Train data for ScanNet v2: 3dmv_scannet_v2_train.zip (6.2G)
- 2D train images can be processed from the ScanNet dataset using the 2d data preparation script in prepare_data
- Expected file structure for 2D data:
scene0000_00/ |--color/ |--[framenum].jpg ⋮ |--depth/ |--[framenum].png (16-bit pngs) ⋮ |--pose/ |--[framenum].txt (4x4 rigid transform as txt file) ⋮ |--label/ (if applicable) |--[framenum].png (8-bit pngs) ⋮ scene0000_01/ ⋮
- Test scenes for ScanNet v2: 3dmv_scannet_v2_test_scenes.zip (110M)
Citation:
If you find our work useful in your research, please consider citing:
@inproceedings{dai20183dmv,
author = {Dai, Angela and Nie{\ss}ner, Matthias},
booktitle = {Proceedings of the European Conference on Computer Vision ({ECCV})},
title = {3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation},
year = {2018}
}
Contact:
If you have any questions, please email Angela Dai at [email protected].
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].