The first workshop on Scene Graph for Structured Intelligence - WACV 2026
March 06, 2026
JW Marriott Starpass in Tucson, Arizona
- Call for Participation
- Learn More from the Workshop Website: https://scene-graph.github.io/SG4SI-WACV26/
- Paper Submission: https://openreview.net/group?id=thecvf.com/WACV/2026/Workshop/SG4SI
======================
1. Workshop Description
Scene graphs provide a structured and interpretable representation of objects, attributes, and relationships in 2D, 3D, and even 4D scenes, serving as a vital bridge between raw visual data and high-level reasoning, which is critical for tasks such as visual reasoning, navigation, and embodied AI. With the rapid rise of multimodal foundation models, integrating scene graphs has become a timely and essential task, offering controllability, explainability, and stronger generalization across different domains and modalities.
This workshop will highlight the latest advances in scene graph generation, representation learning, and their applications in vision–language reasoning, multimodal generation, and robotics. We aim to establish new benchmarks, foster interdisciplinary collaboration, and chart future directions toward the development of structured multimodal intelligence. By uniting researchers from computer vision, NLP, and robotics, the workshop will stimulate impactful discussions and accelerate progress toward trustworthy, general-purpose AI systems.
======================================
2. Call for Papers
We invite original research contributions in (but not limited to) the following areas:
• Scene Graph Generation, methods for constructing scene graphs from images, videos, 3D/4D, and panoptic scenes, including structured modeling, long-tail relation handling, and multi-modal fusion.
• Scene Representation Learning, advances in learning unified, scalable, and transferable scene representations, covering self-supervised learning, contrastive approaches, cross-modal alignment, and integration with large language models (LLMs) and multimodal large language models (MLLMs).
• Scene Graph for Downstream Applications, leveraging structured scene representations for tasks such as image / video generation, visual question answering, spatial reasoning, robotic navigation, ...
• Scene Graph Benchmark and Evaluation, design of new benchmarks, and metrics for scene graph quality, cross-modal grounding, reasoning ability, controllability, and generalization across domains.
=========================
3. Paper Submission
We welcome the submission of long, short, and demo papers. Long papers may include up to 8 pages of content (excluding references), while short and demo papers may include up to 4 pages plus references. Authors will be asked to indicate the paper type and whether they wish their submission to appear in the workshop proceedings; we particularly encourage long papers to be included. All papers must be written in English, follow the double-blind review process, and conform to the WACV formatting guidelines (templates available on the WACV website). Papers should be submitted via the OpenReview system.
=========================
4. Important Date
December 15, 2025, PST, Paper Submission Deadline
January 4, 2026, PST, Notification
January 8, 2026, PST, Camera-ready
March 06, 2026, PST, Workshop Date