View Selection for 3D Captioning via Diffusion Ranking

• URL: http://arxiv.org/abs/2404.07984v1
• Date: Thu, 11 Apr 2024 17:58:11 GMT
• Title: View Selection for 3D Captioning via Diffusion Ranking
• Authors: Tiange Luo, Justin Johnson, Honglak Lee,
• Abstract summary: Cap3D method renders 3D objects into 2D views for captioning using pre-trained models. Some rendered views of 3D objects are atypical, deviating from the training data of standard image captioning models and causing hallucinations. 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.
• Score: 54.78058803763221
• License: http://creativecommons.org/licenses/by/4.0/
• 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.

Related papers

• Open-Vocabulary 3D Semantic Segmentation with Text-to-Image Diffusion Models [57.37244894146089]
  We propose Diff2Scene, which leverages frozen representations from text-image generative models, along with salient-aware and geometric-aware masks, for open-vocabulary 3D semantic segmentation and visual grounding tasks. We show that it outperforms competitive baselines and achieves significant improvements over state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-07-18T16:20:56Z)
• Visual Fact Checker: Enabling High-Fidelity Detailed Caption Generation [34.45033554641476]
  Existing automatic captioning methods for visual content face challenges such as lack of detail, hallucination content, and poor instruction following. We propose VisualFactChecker (VFC), a flexible training-free pipeline that generates high-fidelity and detailed captions for both 2D images and 3D objects. VFC consists of three steps: 1) proposal, where image-to-text captioning models propose multiple initial captions; 2) verification, where a large language model (LLM) utilizes tools such as object detection and VQA models to fact-check proposed captions.
  arXiv  Detail & Related papers  (2024-04-30T17:55:27Z)
• Cross-Modal Self-Training: Aligning Images and Pointclouds to Learn Classification without Labels [69.55622471172941]
  Large-scale vision 2D vision language models, such as CLIP can be aligned with a 3D encoder to learn generalizable (open-vocabulary) 3D vision models. We propose an optimization framework: Cross-MoST: Cross-Modal Self-Training, to improve the label-free classification performance of a zero-shot 3D vision model.
  arXiv  Detail & Related papers  (2024-04-15T21:30:50Z)
• Weakly-Supervised 3D Scene Graph Generation via Visual-Linguistic Assisted Pseudo-labeling [9.440800948514449]
  We propose a weakly-supervised 3D scene graph generation method via Visual-Linguistic Assisted Pseudo-labeling. Our 3D-VLAP exploits the superior ability of current large-scale visual-linguistic models to align the semantics between texts and 2D images. We design an edge self-attention based graph neural network to generate scene graphs of 3D point cloud scenes.
  arXiv  Detail & Related papers  (2024-04-03T07:30:09Z)
• Scalable 3D Captioning with Pretrained Models [63.16604472745202]
  Cap3D is an automatic approach for generating descriptive text for 3D objects. We apply Cap3D to the recently introduced large-scale 3D dataset. Our evaluation, conducted using 41k human annotations from the same dataset, demonstrates that Cap3D surpasses human descriptions in terms of quality, cost, and speed.
  arXiv  Detail & Related papers  (2023-06-12T17:59:03Z)
• CLIP-Guided Vision-Language Pre-training for Question Answering in 3D Scenes [68.61199623705096]
  We design a novel 3D pre-training Vision-Language method that helps a model learn semantically meaningful and transferable 3D scene point cloud representations. We inject the representational power of the popular CLIP model into our 3D encoder by aligning the encoded 3D scene features with the corresponding 2D image and text embeddings. We evaluate our model's 3D world reasoning capability on the downstream task of 3D Visual Question Answering.
  arXiv  Detail & Related papers  (2023-04-12T16:52:29Z)
• 3D-TOGO: Towards Text-Guided Cross-Category 3D Object Generation [107.46972849241168]
  3D-TOGO model generates 3D objects in the form of the neural radiance field with good texture. Experiments on the largest 3D object dataset (i.e., ABO) are conducted to verify that 3D-TOGO can better generate high-quality 3D objects.
  arXiv  Detail & Related papers  (2022-12-02T11:31:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.