CoDi: Low-Shot Counting with Exemplar-Conditioned Diffusion Models

This research explores a novel application of diffusion models for low-shot object counting, a challenging computer vision task. The paper's strength lies in demonstrating the effectiveness of exemplar conditioning, allowing the model to learn from limited examples.

• •Proposes a novel approach to low-shot object counting using diffusion models.
• •Employs exemplar conditioning to enable learning from limited data.
• •Represents an advancement in applying diffusion models to the counting task.