시멘틱 기법은 데이터 종류에 따라, 2차원 이미지, 3차원 포인트 클라우드로 구분할 수 있다.
포인트 클라우드 시멘틱 세그먼테이션의 최신 기법은 다음과 같다.
- RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds (CVPR 2020)
- RangeNet++ and SUMA
- KPConv: Flexible and Deformable Convolution for Point Clouds (2019)
- PointCNN: Convolution On X-Transformed Points (2019)
- Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds
- Exploring Spatial Context for 3D Semantic Segmentation of Point Clouds
- Awesome point cloud analysis
- Associatively Segmenting Instances and Semantics in Point Clouds
- TANet
- Complex-YOLO
- Geometry Sharing Network for 3D Point Cloud Classification and Segmentation. AAAI 2020
- Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data
- PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation
- Semantic Segmentation Editor
- PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
- Dynamic Graph Convolutional Neural Network
- PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
이미지 기반 시멘틱 세그먼테이션의 최신 기법은 다음과 같다.
- MIT Deep Learning
- Deep Learning for Tracking and Detection
- SLAM Implementation: Bundle Adjustment with g2o
- 2019 Guide to Semantic Segmentation
- 2D 3D semantic segmentation
- MIT Driving Scene Segmentation
- FCN (Berkeley version)
- Keras FCN
- DeepLab (CoLab)
- DeepLab for video
Semantic segmentation mAP, FPS
레퍼런스
댓글 없음:
댓글 쓰기