# 3D Segmentation Using Viewpoint-Dependent Spatial Relationships

arXiv: https://arxiv.org/abs/2605.15708 · HTML: https://mlsanity.com/paper/2605.15708/

## Key facts

- Cited by 1 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 24 of its 29 references that are in the map.
- First posted on arXiv 15 May 2026.

## 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, 24 of 24)

- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [PointCLIP: Point Cloud Understanding by CLIP](https://mlsanity.com/paper/2112.02413/) (2021)
- [ScanRefer: 3D Object Localization in RGB-D Scans using Natural Language](https://mlsanity.com/paper/1912.08830/) (2019)
- [ScanQA: 3D Question Answering for Spatial Scene Understanding](https://mlsanity.com/paper/2112.10482/) (2021)
- [Visual Relationship Detection with Language Priors](https://mlsanity.com/paper/1608.00187/) (2016)
- [SQA3D: Situated Question Answering in 3D Scenes](https://mlsanity.com/paper/2210.07474/) (2022)
- [Scan2Cap: Context-aware Dense Captioning in RGB-D Scans](https://mlsanity.com/paper/2012.02206/) (2020)
- [LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planning](https://mlsanity.com/paper/2311.18651/) (2023)
- [Learning 3D Semantic Scene Graphs from 3D Indoor Reconstructions](https://mlsanity.com/paper/2004.03967/) (2020)
- [Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifiers](https://mlsanity.com/paper/2312.08168/) (2023)
- [Multi3DRefer: Grounding Text Description to Multiple 3D Objects](https://mlsanity.com/paper/2309.05251/) (2023)
- [RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics](https://mlsanity.com/paper/2411.16537/) (2024)
- [3D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformer](https://mlsanity.com/paper/2501.01163/) (2025)
- [SpatialLM: Training Large Language Models for Structured Indoor Modeling](https://mlsanity.com/paper/2506.07491/) (2025)
- [MiniGPT-3D: Efficiently Aligning 3D Point Clouds with Large Language Models using 2D Priors](https://mlsanity.com/paper/2405.01413/) (2024)
- [3D-GRAND: A Million-Scale Dataset for 3D-LLMs with Better Grounding and Less Hallucination](https://mlsanity.com/paper/2406.05132/) (2024)
- [SegPoint: Segment Any Point Cloud via Large Language Model](https://mlsanity.com/paper/2407.13761/) (2024)
- [Situational Awareness Matters in 3D Vision Language Reasoning](https://mlsanity.com/paper/2406.07544/) (2024)
- [Reason3D: Searching and Reasoning 3D Segmentation via Large Language Model](https://mlsanity.com/paper/2405.17427/) (2024)
- [Multimodal Referring Segmentation: A Survey](https://mlsanity.com/paper/2508.00265/) (2025)
- [ReferSplat: Referring Segmentation in 3D Gaussian Splatting](https://mlsanity.com/paper/2508.08252/) (2025)
- [3D-GRES: Generalized 3D Referring Expression Segmentation](https://mlsanity.com/paper/2407.20664/) (2024)
- [SURPRISE3D: A Dataset for Spatial Understanding and Reasoning in Complex 3D Scenes](https://mlsanity.com/paper/2507.07781/) (2025)

## Built on by (most cited first, 1 of 1)

- [DSSR-3D: Decoupled Reasoning for View-Dependent Referring in 3D Gaussians](https://mlsanity.com/paper/2610.00040/) (2026)

## Abstract (arXiv)

Recent advances in 3D datasets and multimodal models have greatly improved natural language 3D scene understanding. However, most 3D referring segmentation methods do not explicitly represent the observer viewpoint, making spatial relations such as "left," "right," "front," and "behind" ambiguous and difficult to evaluate. We introduce a viewpoint-aware 3D referring segmentation dataset containing 220k benchmark samples, and scalable to tens of millions of viewpoint-conditioned samples through dense viewpoint sampling. In this dataset, target objects can only be identified through observer-centric spatial relations, making viewpoint-conditioned grounding necessary. We construct the benchmark by leveraging camera poses to automatically annotate observer-centric relations (left/right, front/behind) together with viewpoint-independent relations (above/under). Using this benchmark, we evaluate several existing 3D large multimodal models in a zero-shot setting and find that current models struggle with viewpoint-dependent spatial instructions. We further study how explicit viewpoint information can be incorporated into 3D large multimodal models. We introduce a viewpoint representation that encodes camera poses and conditions the model on the observation viewpoint, improving segmentation accuracy on viewpoint-dependent relations and increasing mIoU from 0.30 to 0.47 compared to a model without viewpoint conditioning. The dataset, code, and trained models will be made publicly available upon acceptance.