Notice: This is a reimplementation of LoGeR; complete code and models will be released upon approval.

LoGeR processes long video streams in chunks with a hybrid memory design to improve large-scale geometric reconstruction quality and consistency.

LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory Junyi Zhang, Charles Herrmann, Junhwa Hur, Chen Sun, Ming-Hsuan Yang, Forrester Cole, Trevor Darrell, Deqing Sun | [Project Webpage] | [arXiv]

git clone https://github.com/junyi42/LoGeR
cd LoGeR
conda create -n loger python=3.11 cmake=3.14.0
conda activate loger
pip install -r requirements.txt

Checkpoints are hosted on Hugging Face:

Please place files as:

• ckpts/LoGeR/latest.pt
• ckpts/LoGeR_star/latest.pt

Example commands:

wget -O ckpts/LoGeR/latest.pt "https://huggingface.co/Junyi42/LoGeR/resolve/main/LoGeR/latest.pt?download=true"
wget -O ckpts/LoGeR_star/latest.pt "https://huggingface.co/Junyi42/LoGeR/resolve/main/LoGeR_star/latest.pt?download=true"

For demo usage, please directly refer to:

• demo_run.sh

For evaluation instructions, please refer to:

• eval/eval.md

If you find our work useful, please cite:

@article{zhang2026loger,
  title={LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory},
  author={Zhang, Junyi and Herrmann, Charles and Hur, Junhwa and Sun, Chen and Yang, Ming-Hsuan and Cole, Forrester and Darrell, Trevor and Sun, Deqing},
  journal={arXiv preprint arXiv:2603.03269},
  year={2026}
}

Our code is based on Pi3 and LaCT, our camera pose estimation evaluation script is based on TTT3R & VBR, and our visualization code is based on Viser. We thank the authors for their excellent work!