3D Instance Segmentation
59 papers with code • 8 benchmarks • 13 datasets
Image: OccuSeg
Benchmarks
These leaderboards are used to track progress in 3D Instance Segmentation
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Libraries
Use these libraries to find 3D Instance Segmentation models and implementations
ScanNet
S3DIS
KITTI-360
PartNet
SceneNN
STPLS3D
ScanNet200
CREMI
FOR-instance
Latest papers
AnyStar: Domain randomized universal star-convex 3D instance segmentation
neel-dey/anystar
•
•
Star-convex shapes arise across bio-microscopy and radiology in the form of nuclei, nodules, metastases, and other units.
OpenMask3D: Open-Vocabulary 3D Instance Segmentation
OpenMask3D/openmask3d
•
•
NeurIPS 2023
In this work, we address this limitation, and propose OpenMask3D, which is a zero-shot approach for open-vocabulary 3D instance segmentation.
Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Geometry and color information provided by the point clouds are both crucial for 3D scene understanding.
Fully Sparse Fusion for 3D Object Detection
bravegroup/fullysparsefusion
•
•
In this paper, we study how to effectively leverage image modality in the emerging fully sparse architecture.
Instance Neural Radiance Field
lyclyc52/instance_nerf
•
•
ICCV 2023
This paper presents one of the first learning-based NeRF 3D instance segmentation pipelines, dubbed as {\bf \inerflong}, or \inerf.
You Only Need One Thing One Click: Self-Training for Weakly Supervised 3D Scene Understanding
liuzhengzhe/One-Thing-One-Click
•
•
3D scene understanding, e. g., point cloud semantic and instance segmentation, often requires large-scale annotated training data, but clearly, point-wise labels are too tedious to prepare.
ISBNet: a 3D Point Cloud Instance Segmentation Network with Instance-aware Sampling and Box-aware Dynamic Convolution
VinAIResearch/ISBNet
•
•
CVPR 2023
Existing 3D instance segmentation methods are predominated by the bottom-up design -- manually fine-tuned algorithm to group points into clusters followed by a refinement network.
Top-Down Beats Bottom-Up in 3D Instance Segmentation
samsunglabs/td3d
•
•
Most 3D instance segmentation methods exploit a bottom-up strategy, typically including resource-exhaustive post-processing.
Superpoint Transformer for 3D Scene Instance Segmentation
The key step in this framework is a novel query decoder with transformers that can capture the instance information through the superpoint cross-attention mechanism and generate the superpoint masks of the instances.
GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Parts
PKU-EPIC/GAPartNet
•
•
CVPR 2023
Based on GAPartNet, we investigate three cross-category tasks: part segmentation, part pose estimation, and part-based object manipulation.
