# CLIP-Guided Vision-Language Pre-training for Question Answering in 3D Scenes

arXiv: https://arxiv.org/abs/2304.06061 · HTML: https://mlsanity.com/paper/2304.06061/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 26 of its 31 references that are in the map.
- First posted on arXiv 12 Apr 2023.

## Placed in

- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks](https://mlsanity.com/paper/2004.06165/) (2020)
- [Open3D: A Modern Library for 3D Data Processing](https://mlsanity.com/paper/1801.09847/) (2018)
- [Deep Hough Voting for 3D Object Detection in Point Clouds](https://mlsanity.com/paper/1904.09664/) (2019)
- [From Recognition to Cognition: Visual Commonsense Reasoning](https://mlsanity.com/paper/1811.10830/) (2018)
- [PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding](https://mlsanity.com/paper/2007.10985/) (2020)
- [ScanRefer: 3D Object Localization in RGB-D Scans using Natural Language](https://mlsanity.com/paper/1912.08830/) (2019)
- [MotionCLIP: Exposing Human Motion Generation to CLIP Space](https://mlsanity.com/paper/2203.08063/) (2022)
- [ScanQA: 3D Question Answering for Spatial Scene Understanding](https://mlsanity.com/paper/2112.10482/) (2021)
- [Deep Modular Co-Attention Networks for Visual Question Answering](https://mlsanity.com/paper/1906.10770/) (2019)
- [Embodied Question Answering](https://mlsanity.com/paper/1711.11543/) (2017)
- [An End-to-End Transformer Model for 3D Object Detection](https://mlsanity.com/paper/2109.08141/) (2021)
- [SQA3D: Situated Question Answering in 3D Scenes](https://mlsanity.com/paper/2210.07474/) (2022)
- [Language-Grounded Indoor 3D Semantic Segmentation in the Wild](https://mlsanity.com/paper/2204.07761/) (2022)
- [Self-Supervised Pretraining of 3D Features on any Point-Cloud](https://mlsanity.com/paper/2101.02691/) (2021)
- [CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding](https://mlsanity.com/paper/2203.00680/) (2022)
- [SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations](https://mlsanity.com/paper/2112.04680/) (2021)
- [Language Grounding with 3D Objects](https://mlsanity.com/paper/2107.12514/) (2021)
- [Spatiality-guided Transformer for 3D Dense Captioning on Point Clouds](https://mlsanity.com/paper/2204.10688/) (2022)
- [Learning from 2D: Contrastive Pixel-to-Point Knowledge Transfer for 3D Pretraining](https://mlsanity.com/paper/2104.04687/) (2021)

## Built on by (most cited first, 30 of 55)

- [LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planning](https://mlsanity.com/paper/2311.18651/) (2023)
- [Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifiers](https://mlsanity.com/paper/2312.08168/) (2023)
- [Chat-3D: Data-efficiently Tuning Large Language Model for Universal Dialogue of 3D Scenes](https://mlsanity.com/paper/2308.08769/) (2023)
- [Grounded 3D-LLM with Referent Tokens](https://mlsanity.com/paper/2405.10370/) (2024)
- [LiDAR-LLM: Exploring the Potential of Large Language Models for 3D LiDAR Understanding](https://mlsanity.com/paper/2312.14074/) (2023)
- [FG-CLIP: Fine-Grained Visual and Textual Alignment](https://mlsanity.com/paper/2505.05071/) (2025)
- [3D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformer](https://mlsanity.com/paper/2501.01163/) (2025)
- [Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces](https://mlsanity.com/paper/2503.19199/) (2025)
- [Inst3D-LMM: Instance-Aware 3D Scene Understanding with Multi-modal Instruction Tuning](https://mlsanity.com/paper/2503.00513/) (2025)
- [Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversion](https://mlsanity.com/paper/2502.04263/) (2025)
- [MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations](https://mlsanity.com/paper/2406.09401/) (2024)
- [M3DBench: Let's Instruct Large Models with Multi-modal 3D Prompts](https://mlsanity.com/paper/2312.10763/) (2023)
- [A Survey on Text-guided 3D Visual Grounding: Elements, Recent Advances, and Future Directions](https://mlsanity.com/paper/2406.05785/) (2024)
- [Text-Guided Attention is All You Need for Zero-Shot Robustness in Vision-Language Models](https://mlsanity.com/paper/2410.21802/) (2024)
- [Bridging the Gap between 2D and 3D Visual Question Answering: A Fusion Approach for 3D VQA](https://mlsanity.com/paper/2402.15933/) (2024)
- [3D Vision and Language Pretraining with Large-Scale Synthetic Data](https://mlsanity.com/paper/2407.06084/) (2024)
- [Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning](https://mlsanity.com/paper/2402.17251/) (2024)
- [Select and Distill: Selective Dual-Teacher Knowledge Transfer for Continual Learning on Vision-Language Models](https://mlsanity.com/paper/2403.09296/) (2024)
- [Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation](https://mlsanity.com/paper/2407.03056/) (2024)
- [Open-Vocabulary Calibration for Fine-tuned CLIP](https://mlsanity.com/paper/2402.04655/) (2024)
- [What do we learn from inverting CLIP models?](https://mlsanity.com/paper/2403.02580/) (2024)
- [RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement](https://mlsanity.com/paper/2404.01889/) (2024)
- [Embodied Intelligence for 3D Understanding: A Survey on 3D Scene Question Answering](https://mlsanity.com/paper/2502.00342/) (2025)
- [3D Question Answering for City Scene Understanding](https://mlsanity.com/paper/2407.17398/) (2024)
- [External Knowledge Enhanced 3D Scene Generation from Sketch](https://mlsanity.com/paper/2403.14121/) (2024)
- [Understanding and Mitigating Miscalibration in Prompt Tuning for Vision-Language Models](https://mlsanity.com/paper/2410.02681/) (2024)
- [PerLA: Perceptive 3D Language Assistant](https://mlsanity.com/paper/2411.19774/) (2024)
- [Probabilistic Embeddings for Frozen Vision-Language Models: Uncertainty Quantification with Gaussian Process Latent Variable Models](https://mlsanity.com/paper/2505.05163/) (2025)
- [Multimodal 3D Reasoning Segmentation with Complex Scenes](https://mlsanity.com/paper/2411.13927/) (2024)
- [HIS-GPT: Towards 3D Human-In-Scene Multimodal Understanding](https://mlsanity.com/paper/2503.12955/) (2025)

## Abstract (arXiv)

Training models to apply linguistic knowledge and visual concepts from 2D images to 3D world understanding is a promising direction that researchers have only recently started to explore. In this work, 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 produced by CLIP. To assess our model's 3D world reasoning capability, we evaluate it on the downstream task of 3D Visual Question Answering. Experimental quantitative and qualitative results show that our pre-training method outperforms state-of-the-art works in this task and leads to an interpretable representation of 3D scene features.