Description: Computer vision, as a branch of computer science, has undergone dynamic changes due to the emergence of AI and large models based on various neural structures. Development of large-scale neural models, including foundation models, vision transformers, and vision–language architectures has substantially changed computer vision research. The goal of this workshop on Computer Vision in the Era of Large Neural Models (CV-LNM) is to invite and bring together scientists who create or use modern CV methods in their research.

Scope of the workshop includes, but is not limited, to the following topics:

- Foundation models trained on large data sets for CV
- Large pretrained vision models and vision transformers
- Vision–language and multimodal learning
- Few-shot and zero-shot learning in CV
- Novel transformer architectures
- Content based image/video/multimedia retrieval
- Large-scale models for spatio-temporal perception, including video understanding, 3D -reconstruction, point-cloud processing, and visual perception
- CV and large neural models in medicine
- CV based on large neural models in robotics
- Industrial applications of CV in the era of large neural models
- Parameter-efficient fine-tuning and domain adaptation
- Data augmentation, synthetic data, semi- and self-supervised learning
- Intelligent multi-modal agents
- Implementation and practical issues: large model training techniques, hardware -implementations (GPU, ASIC, FPGA), cloud computations etc.
- Visualization of high dimensional data
- Model interpretability and prediction explainability
- Diffusion and generative based models in CV
- Efficiency, optimization, and deployment in real-world systems