LRM: Large Reconstruction Model
for Single Image to 3D

Anonymous. Paper Under Review.
Create high-fidelity 3D object mesh from a single image in 5 SECONDS.
All testing images shown below are never seen by the model in training.

Phone Camera Captured

Generated Images (Adobe Firefly)



Interactable Meshes

Toy Giraffe

Tea Cup

Teddy Bear

Wood Peafowl


Toy Flower


We propose the first Large Reconstruction Model (LRM) that predicts the 3D model of an object from a single input image within just 5 seconds. In contrast to many previous methods that are trained on small-scale datasets such as ShapeNet in a category-specific fashion, LRM adopts a highly scalable transformer-based architecture with 500 million learnable parameters to directly predict a neural radiance field (NeRF) from the input image. We train our model in an end-to-end manner on massive multi-view data containing around 1 million objects, including both synthetic renderings from Objaverse and real captures from MVImgNet. This combination of a high-capacity model and large-scale training data empowers our model to be highly generalizable and produce high-quality 3D reconstructions from various testing inputs including real-world in-the-wild captures and images from generative models.


Figure 1. The overall architecture of LRM, a fully-differentiable transformer-based encoedr-decoder framework for single-image to NeRF reconstruction. LRM applies a pre-trained vision model (DINO) to encode the input image, where the image features are projected to a 3D triplane representation by a large transformer decoder via cross-attention, followed by a multi-layer perceptron to predict the point color and density for volumetric rendering. The entire network is trained end-to-end on around a million of 3D data by simply minimizing the difference between the rendered images and ground truth images at novel views.

More Results

Phone Camera Captured



Google Scanned Objects

Amazon Berkeley Objects



    author    = {Anonymous},
    title     = {LRM: Large Reconstruction Model for Single Image to 3D},
    journal   = {Under Review},
    year      = {2023},