Diffusion models have achieved remarkable progress in the field of image generation due to their outstanding capabilities. However, these models require substantial computing resources because of the multi-step denoising process during inference. While traditional pruning methods have been employed to optimize these models, the retraining process necessitates large-scale training datasets and extensive computational costs to maintain generalization ability, making it neither convenient nor efficient. Recent studies attempt to utilize the similarity of features across adjacent denoising stages to reduce computational costs through simple and static strategies. However, these strategies cannot fully harness the potential of the similar feature patterns across adjacent timesteps. In this work, we propose a novel pruning method that derives an efficient diffusion model via a more intelligent and differentiable pruner. At the core of our approach is casting the model pruning process into a SubNet search process. Specifically, we first introduce a SuperNet based on standard diffusion via adding some backup connections built upon the similar features. We then construct a plugin pruner network and design optimization losses to identify redundant computation. Finally, our method can identify an optimal SubNet through few-step gradient optimization and a simple post-processing procedure. We conduct extensive experiments on various diffusion models including Stable Diffusion series and DiTs. Our DiP-GO approach achieves 4.4 x speedup for SD-1.5 without any loss of accuracy, significantly outperforming the previous state-of-the-art methods.

In this paper, a methodology for the detection of buried mines in airborne multispectral imagery is explored. The approach is based on utilizing the color texture information in the buried mine signatures, which is extracted using the cross-co-occurrence texture features. A systematic two-stage approach, using Bhattacharya coefficient-based analysis and principal feature analysis, is developed for the selection of a small subset of discriminatory features. Detection results from actual airborne data from two different sites are presented. The performances are compiled for four different feature-based detectors, and are compared with the conventional multiband RX anomaly detector, to validate the feature selection approach and demonstrate buried mine detection performance based on texture features.