DetDiffusion: Synergizing Generative and Perceptive Models for Enhanced Data Generation and Perception

arXiv:2403.13304v1 Announce Type: new
Abstract: Current perceptive models heavily depend on resource-intensive datasets, prompting the need for innovative solutions. Leveraging recent advances in diffusion models, synthetic data, by constructing image inputs from various annotations, proves beneficial for downstream tasks. While prior methods have separately addressed generative and perceptive models, DetDiffusion, for the first time, harmonizes both, tackling the challenges in generating effective data for perceptive models. To enhance image generation with perceptive models, we introduce perception-aware loss (P.A. loss) through …

abstract advances annotations arxiv cs.cv current data datasets diffusion diffusion models generative image inputs perception prior prompting solutions synthetic synthetic data tasks type