Papers › GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models

GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models

2 Jan 2025arXiv:2501.01428archive 2025-07-28

Zhangyang Qi, Zhixiong Zhang, Ye Fang, Jiaqi Wang, Hengshuang Zhao

In recent years, 2D Vision-Language Models (VLMs) have made significant strides in image-text understanding tasks. However, their performance in 3D spatial comprehension, which is critical for embodied intelligence, remains limited. Recent advances have leveraged 3D point clouds and multi-view images as inputs, yielding promising results. However, we propose exploring a purely vision-based solution inspired by human perception, which merely relies on visual cues for 3D spatial understanding. This paper empirically investigates the limitations of VLMs in 3D spatial knowledge, revealing that their primary shortcoming lies in the lack of global-local correspondence between the scene and individual frames. To address this, we introduce GPT4Scene, a novel visual prompting paradigm in VLM training and inference that helps build the global-local relationship, significantly improving the 3D spatial understanding of indoor scenes. Specifically, GPT4Scene constructs a 3D Bird's Eye View (BEV) image from the video and marks consistent object IDs across both frames and the BEV image. The model then inputs the concatenated BEV image and video frames with markers. In zero-shot evaluations, GPT4Scene improves performance over closed-source VLMs like GPT-4o. Additionally, we prepare a processed video dataset consisting of 165K text annotation to fine-tune open-source VLMs, achieving state-of-the-art performance on all 3D understanding tasks. Surprisingly, after training with the GPT4Scene paradigm, VLMs consistently improve during inference, even without visual prompting and BEV image as explicit correspondence. It demonstrates that the proposed paradigm helps VLMs develop an intrinsic ability to understand 3D scenes, which paves the way for a noninvasive approach to extending pre-trained VLMs for 3D scene understanding.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 8 code samples harvested from 1 repository linked to this paper: 8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 8 samples from 1 repository, 8 ran (8 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Qi-Zhangyang/GPT4Scene officialmentioned on GitHubpytorchApache-2.0Syntology: 8 samples, 8 ran (0 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

8 samples harvested: 8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

8ran

Licence: 0 of the 8 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Qi-Zhangyang/GPT4Scene. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

Qi-Zhangyang/GPT4Scene official repository 8 samples, 8 ran (0 instrument-failed) · 0 unverified
add_step_prompts VLN-Ego-making/VLN-Ego/r2r_rxr_dataset_convert_sft.py ran Apache-2.0 (permissive) · 5064009b43a00726 · report fixture: signature navigation_data:dict/0/?, num_actions:int/0/int · returned dict · level recorded 2026-09-24
bulid_message evaluate/qwen2_vl.py ran Apache-2.0 (permissive) · 7173826fa5a3b3ec · report fixture: signature prompt:str/0/str, config:MinimalConfig/0/?, eval_name:str/0/str, bev_mf_paths:list/1/str · returned list · level recorded 2026-09-24
create_dataset in 47 papers evaluate/val_dataset.py ran Apache-2.0 (permissive) · d1cc4570ba8bbd09 · report fixture: signature config:Config/0/? · returned list · level recorded 2026-09-24
create_loader in 6 papers evaluate/val_dataset.py ran Apache-2.0 (permissive) · c11a23c56c87bd2d · report fixture: signature datasets:list/1/DummyDataset, samplers:list/1/NoneType, batch_size:list/1/int, num_workers:list/1/int, is_trains:list/1/bool, collate_fns:list/1/callable · returned list · level recorded 2026-09-24
create_navigation_data VLN-Ego-making/VLN-Ego/r2r_rxr_dataset_convert_sft.py ran Apache-2.0 (permissive) · 13da697eeaf18b88 · report fixture: signature base_folder:str/0/str, max_frames:int/0/int, num_actions:int/0/int · returned dict · level recorded 2026-09-24
create_sampler in 14 papers evaluate/val_dataset.py ran Apache-2.0 (permissive) · 8d7bf7ad0e60457f · report fixture: signature datasets:list/1/DummyDataset, shuffles:list/1/bool, num_tasks:int/0/int, global_rank:int/0/int · returned list · level recorded 2026-09-22
get_step_images VLN-Ego-making/VLN-Ego/r2r_rxr_dataset_convert_sft.py ran Apache-2.0 (permissive) · 10db9373c4cbf700 · report fixture: signature ep_folder:str/0/str · returned list · level recorded 2026-09-25
preprocess_image_nearest evaluate/qwen2_vl.py ran Apache-2.0 (permissive) · 174c802b9d2bb222 · report fixture: signature image_path:str/0/str, image_resolution:int/0/int, perform_resize:bool/0/bool · returned tuple · level recorded 2026-09-24

Tasks

Scene UnderstandingVisual Promptingtext annotation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections