StandardSim
StandardSim is a large-scale photorealistic synthetic dataset for retail environments with over 25k examples. Modalities include rgb, depth, before/after, instance segmentation labels, depth, and normals.
Project Page: https://standard-ai.github.io/Standard-Sim/
ICIAP 2022 Paper: https://arxiv.org/abs/2202.02418
Data: https://drive.google.com/drive/folders/1KRjuL1wxe2VqYKKIOAKa0-FXjTNgr4TX
Code for training models coming soon.
Change Detection Task
Change Detection Labels and Model Predictions
Data Modalities
Dataset Comparison
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This license is inclusive of both this codebase and the StandardSim data release.