Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Multimodal Models
Matt Deitke, Chirstopher Clark, et al.
Technical report 2024

A family of open multimodal models that are as good as the best commerical ones. My contribution was creating a novel evaluation benchmark for testing fine-grained world knowledge in VQA.

• Arxiv
• Page
• Code
• BibTex

Task Me Anything
Jieyu Zhang, Weikai Huang, Zixian Ma, Oscar Michel, Dong He, Tanmay Gupta, Wei-Chu Ma, Ali Farhadi, Aniruddha Kembhavi, Ranjay Krishna
NeurIPS Datasets & Benchmarks 2024

A new way of thinking about how we can evaluate models using synthetic visual data.

• Paper
• Page
• Models
• Demo

OBJect 3DIT: Language-guided 3D-aware Image Editing
Oscar Michel, Anand Bhattad, Eli VanderBilt, Ranjay Krishna, Ani Kembhavi, Tanmay Gupta
NeurIPS 2023

Using synthetic data to give diffusion models intutitive controls that enable geometrically accurate 3D manipulations of objects in images.

• Arxiv
• Page
• Code
• BibTex

Objaverse-XL: A Universe of 10M+ 3D Objects
Matt Deitke, Ruoshi Liu, Matthew Wallingford, Huong Ngo, Oscar Michel, Aditya Kusupati, Alan Fan, Christian Laforte, Vikram Voleti, Samir Yitzhak Gadr, Eli VanderBilt, Aniruddha Kembhavi, Carl Vondrick, Georgia Gkioxari, Kiana Ehsani, Ludwig Schmidt, Ali Farhadi
NeurIPS Datasets & Benchmarks 2023

A dataset 10x the scale of Objaverse 1.0 and a 3D object reconstruction foundation model.

• Arxiv
• Page
• BibTex

Objaverse: A Universe of Annotated 3D Objects
Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, Ali Farhadi
CVPR 2023

Objaverse is a large dataset of 800k+ 3D models.

• Arxiv
• Page
• Colab
• BibTex

Text2Mesh: Text-Driven Neural Stylization for Meshes
Oscar Michel^*, Roi Bar On^*, Richard Liu^*, Sagie Benaim, Rana Hanocka
CVPR 2022 (Oral)

Text2Mesh is an algorithm for language guided stylization of a 3D object.

• Arxiv
• Page
• Code
• BibTex