Pith. sign in

REVIEW 9 cited by

Benchmarking Spatial Relationships in Text-to-Image Generation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2212.10015 v3 pith:TGHXXADA submitted 2022-12-20 cs.CV cs.AIcs.CL

classification cs.CVcs.AIcs.CL
keywords spatialmodelsobjectsrelationshipsunderstandingevaluationvisorability
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Spatial understanding is a fundamental aspect of computer vision and integral for human-level reasoning about images, making it an important component for grounded language understanding. While recent text-to-image synthesis (T2I) models have shown unprecedented improvements in photorealism, it is unclear whether they have reliable spatial understanding capabilities. We investigate the ability of T2I models to generate correct spatial relationships among objects and present VISOR, an evaluation metric that captures how accurately the spatial relationship described in text is generated in the image. To benchmark existing models, we introduce a dataset, $\mathrm{SR}_{2D}$, that contains sentences describing two or more objects and the spatial relationships between them. We construct an automated evaluation pipeline to recognize objects and their spatial relationships, and employ it in a large-scale evaluation of T2I models. Our experiments reveal a surprising finding that, although state-of-the-art T2I models exhibit high image quality, they are severely limited in their ability to generate multiple objects or the specified spatial relations between them. Our analyses demonstrate several biases and artifacts of T2I models such as the difficulty with generating multiple objects, a bias towards generating the first object mentioned, spatially inconsistent outputs for equivalent relationships, and a correlation between object co-occurrence and spatial understanding capabilities. We conduct a human study that shows the alignment between VISOR and human judgement about spatial understanding. We offer the $\mathrm{SR}_{2D}$ dataset and the VISOR metric to the community in support of T2I reasoning research.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Rule-VLN: Bridging Perception and Compliance via Semantic Reasoning and Geometric Rectification

    cs.AI 2026-04 unverdicted novelty 7.0 of 10

    Rule-VLN is the first large-scale benchmark injecting 177 regulatory categories into an urban environment, and the proposed SNRM module equips pre-trained VLN agents with zero-shot semantic reasoning and detour planni...

  2. Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Image-generation models become competitive on spatial benchmarks when answers are expressed as protocol-constrained pixels, while text-output VLMs retain an edge on compositional reasoning — with a parser-sensitivity ...

  3. Transferability Between Understanding and Generation in Unified Multimodal Models

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Cross-task capability transfer in UMMs is architecture-dependent and can be exploited by training understanding to improve generation while avoiding distribution shift.

  4. Human Preference-Aligned Concept Customization Benchmark via Decomposed Evaluation

    cs.CV 2025-09 conditional novelty 6.0 of 10

    D-GPTScore, which averages GPT-4o's per-aspect ratings of concept-customized images, correlates with human preference at 0.78 Pearson on the new CC-AlignBench, beating prior metrics.

  5. Canvas3D: Empowering Precise Spatial Control for Image Generation with Constraints from a 3D Virtual Canvas

    cs.HC 2025-08 conditional novelty 6.0 of 10

    Canvas3D lets users arrange objects in a 3D canvas generated from a text prompt, then feeds depth, skeleton, and lighting constraints to diffusion models to produce images that match the layout.

  6. AMVICC: A Novel Benchmark for Cross-Modal Failure Mode Profiling for VLMs and IGMs

    cs.CV 2026-01 conditional novelty 5.0 of 10

    A cross-modal benchmark derived from MMVP shows VLMs and IGMs share several elementary visual-reasoning failure modes, with IGMs struggling most on explicit attribute-control prompts.

  7. Why Settle for Mid: A Probabilistic Viewpoint to Spatial Relationship Alignment in Text-to-image Models

    cs.CV 2025-06 conditional novelty 5.0 of 10

    A probabilistic overlap measure for object positions yields a human-aligned spatial relationship metric and a training-free generation guidance method for text-to-image models.

  8. Towards Evaluating Robustness of Prompt Adherence in Text to Image Models

    cs.CV 2025-07 conditional novelty 4.0 of 10

    New benchmark results show that Stable Diffusion 3.x and Janus Pro models struggle to place simple geometric shapes in the correct image quadrant, with best F1 scores around 0.41 to 0.5.

  9. Re-Thinking the Automatic Evaluation of Image-Text Alignment in Text-to-Image Models

    cs.CL 2025-06 conditional novelty 4.0 of 10

    Current image-text alignment metrics, including CLIPScore and DSGScore, produce unstable model rankings under random seeds and are highly sensitive to tiny image perturbations.

Pith tools