Fine-Tuning Image-Conditional Diffusion Models is Easier than you Think

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 753–762

Abstract

Recent work showed that large diffusion models can be reused as highly precise monocular depth estimators by casting depth estimation as an image-conditional image generation task. While the proposed model achieved state-of-the-art results, high computational demands due to multi-step inference limited its use in many scenarios. In this paper, we show that the perceived inefficiency was caused by a flaw in the inference pipeline that has so far gone unnoticed. The fixed model performs comparably to the best previously reported configuration while being more than 200x faster. To optimize for downstream task performance, we perform end-to-end fine-tuning on top of the single-step model with task-specific losses and get a deterministic model that outperforms all other diffusion-based depth and normal estimation models on common zero-shot benchmarks. We surprisingly find that this fine-tuning protocol also works directly on Stable Diffusion and achieves comparable performance to current state-of-the-art diffusion-based depth and normal estimation models, calling into question some of the conclusions drawn from prior works.

Authors 5

1. Karim Abou Zeid Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

2. Christian Schmidt Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

3. Daan de Geus Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

4. Alexander Hermans Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

5. Bastian Leibe Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University