Call for Papers

Topics

We invite contributions on the topics related to memory modeling and computer vision. Topics of interest include but are not limited to:

• Hybrid/ state space models for vision applications

• Novel memory-augmented architectures in generative models, foundation models, and diffusion networks

• Dense Associative Memory, Energy Transformers, and other energy-based models for vision applications

• Memory-enhanced learning algorithms, such as continual learning and lifelong learning, with episodic memory in computer vision

• Video and sequential vision with temporal memory

• Retrieval-augmented vision models with external memory

• Multimodal memory-enhanced AI systems

• Foundation models with long-term visual context

• Human memorability and cognitive insights in memory system design

• Scalability and efficiency in memory-enhanced vision model training and inference

• Evaluation, benchmarks, and datasets for deploying memory-augmented vision models

• Memory-based models for addressing ethical, privacy, and safety considerations

Submission Guidelines

All submissions should use the ICCV 2025 paper template.

• Short Papers or Extended Abstracts (non-archival; anonymous submission): Up to 4 pages, excluding references and supplementary materials

• Papers accepted to ICCV (non-archival; non-anonymous): Full-length papers (up to 8 pages, excluding references) accepted to the main ICCV 2025 conference on topics relevant to the workshop.

Submission site: https://openreview.net/group?id=thecvf.com/ICCV/2025/Workshop/MemVis

Submissions are due August 20, 2025, with notifications to authors on August 27, 2025 and camera-ready papers due by September 15, 2025 (all due 11:59 pm AoE)