Openlrm-mix-large-1.1 Open Source Model - Easily Generate 3D Models from a Single Image for Free

Home

Openlrm Mix Large 1.1

Developed by zxhezexin

OpenLRM is an open-source implementation of the LRM paper for generating 3D models from a single image

3D Vision

Transformers

#Single-image 3D Reconstruction #High-resolution Rendering #Tri-plane Decoding

Downloads 98

Release Time : 3/4/2024

Model Overview

OpenLRM is an image-to-3D generative model capable of reconstructing high-quality 3D objects from a single input image. This project is the open-source implementation of the original LRM paper.

Model Features

Multi-scale Model Options

Offers small/base/large model scales to accommodate different computational resource needs

Dual Training Data Versions

Provides both pure Objaverse-trained version and Objaverse+MVImgNet mixed training version

Improved Image Encoder

Uses DINOv2 model with register tokens as the image encoder

Multi-level Tri-plane Decoding

Tri-plane decoder contains 4 levels to enhance reconstruction quality

Model Capabilities

Single-view 3D reconstruction

3D mesh generation

3D shape prediction

Use Cases

3D Content Creation

Game Asset Generation

Quickly generate 3D game models from concept art

Virtual Scene Construction

Create 3D scene elements based on reference images

Industrial Design

Product Prototyping

Generate preliminary 3D models from product photos

🚀 Model Card for OpenLRM V1.1

This model card provides detailed information about the OpenLRM V1.1 project, an open - source implementation of the LRM paper.

🚀 Quick Start

This model card is for the OpenLRM project, an open - source implementation of the paper LRM. The information in this model card corresponds to Version 1.1.

✨ Features

Model Details

Training Data

Property	Details
openlrm-obj-small-1.1	Objaverse
openlrm-obj-base-1.1	Objaverse
openlrm-obj-large-1.1	Objaverse
openlrm-mix-small-1.1	Objaverse + MVImgNet
openlrm-mix-base-1.1	Objaverse + MVImgNet
openlrm-mix-large-1.1	Objaverse + MVImgNet

Model Architecture (version==1.1)

Type	Layers	Feat. Dim	Attn. Heads	Triplane Dim.	Input Res.	Image Encoder	Size
small	12	512	8	32	224	dinov2_vits14_reg	446M
base	12	768	12	48	336	dinov2_vitb14_reg	1.04G
large	16	1024	16	80	448	dinov2_vitb14_reg	1.81G

Training Settings

Type	Rend. Res.	Rend. Patch	Ray Samples
small	192	64	96
base	288	96	96
large	384	128	128

Notable Differences from the Original Paper

We do not use the deferred back - propagation technique in the original paper.
We used random background colors during training.
The image encoder is based on the DINOv2 model with register tokens.
The triplane decoder contains 4 layers in our implementation.

📄 License

The model weights are released under the Creative Commons Attribution - NonCommercial 4.0 International License.
They are provided for research purposes only, and CANNOT be used commercially.

📚 Documentation

Disclaimer

This model is an open - source implementation and is NOT the official release of the original research paper. While it aims to reproduce the original results as faithfully as possible, there may be variations due to model implementation, training data, and other factors.

Ethical Considerations

This model should be used responsibly and ethically, and should not be used for malicious purposes.
Users should be aware of potential biases in the training data.
The model should not be used under the circumstances that could lead to harm or unfair treatment of individuals or groups.

Usage Considerations

The model is provided "as is" without warranty of any kind.
Users are responsible for ensuring that their use complies with all relevant laws and regulations.
The developers and contributors of this model are not liable for any damages or losses arising from the use of this model.

💡 Usage Tip

This model card is subject to updates and modifications. Users are advised to check for the latest version regularly.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご