# View Selection for 3D Captioning via Diffusion Ranking

Source: [View Selection for 3D Captioning via Diffusion Ranking](<https://www.sota2.com/research/paper/68fa7a1b1e209131cea81314>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-31T05:00:03.851Z

Authors: Tiange Luo, Justin Johnson, Honglak Lee

Published: 2024-04-11T17:58:11.000Z

## Abstract

Scalable annotation approaches are crucial for constructing extensive 3D-text datasets, facilitating a broader range of applications. However, existing methods sometimes lead to the generation of hallucinated captions, compromising caption quality. This paper explores the issue of hallucination in 3D object captioning, with a focus on Cap3D method, which renders 3D objects into 2D views for captioning using pre-trained models. We pinpoint a major challenge: certain rendered views of 3D objects are atypical, deviating from the training data of standard image captioning models and causing hallucinations. To tackle this, we present DiffuRank, a method that leverages a pre-trained text-to-3D model to assess the alignment between 3D objects and their 2D rendered views, where the view with high alignment closely represent the object's characteristics. By ranking all rendered views and feeding the top-ranked ones into GPT4-Vision, we enhance the accuracy and detail of captions, enabling the correction of 200k captions in the Cap3D dataset and extending it to 1 million captions across Objaverse and Objaverse-XL datasets. Additionally, we showcase the adaptability of DiffuRank by applying it to pre-trained text-to-image models for a Visual Question Answering task, where it outperforms the CLIP model.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Object Captioning on Cap3D subset of 3186 3D objects (test)](<https://www.sota2.com/research/sota/3d-object-captioning-on-cap3d-subset-of-3186-3d-objects-test.md>): 3D Object Captioning; Cap3D subset of 3186 3D objects (test); CLIP Image-Text Score: 0.291.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2404.07984>)

[PDF](<https://arxiv.org/pdf/2404.07984.pdf>)