OGC: Unsupervised 3D Object Segmentation from Rigid Dynamics of Point Clouds

Authors: Ziyang Song, Bo Yang

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We extensively evaluate our method on five datasets, demonstrating the superior performance for object part instance segmentation and general object segmentation in both indoor and the challenging outdoor scenarios. Our code and data are available at https://github.com/v LAR-group/OGC
Researcher Affiliation Academia v LAR Group, The Hong Kong Polytechnic University ziyang.song@connect.polyu.hk bo.yang@polyu.edu.hk
Pseudocode Yes Algorithm 1 Iterative optimization of object segmentation and scene flow estimation. Assume the whole train split has S point cloud pairs: {(P t, P t+1)1 (P t, P t+1)S}.
Open Source Code Yes Our code and data are available at https://github.com/v LAR-group/OGC
Open Datasets Yes Our code and data are available at https://github.com/v LAR-group/OGC
Dataset Splits Yes In total, we create 3750, 250, and 1000 indoor rooms (scenes) for training/validation/test splits.
Hardware Specification Yes The scene flow network is trained for 200 epochs... using a single NVIDIA A100 GPU.
Software Dependencies Yes Our code is written in Python 3.8 and PyTorch 1.9.
Experiment Setup Yes We train the object segmentation network for 100 epochs, with Adam optimizer (β1=0.9, β2=0.999), an initial learning rate of 0.001 and batch size of 8. The scene flow network is trained for 200 epochs with Adam optimizer, an initial learning rate of 0.001 and batch size of 4.