Hierarchical Dynamic Image Harmonization

16 Nov 2022  ·  Haoxing Chen, Zhangxuan Gu, Yaohui Li, Jun Lan, Changhua Meng, Weiqiang Wang, Huaxiong Li ·

Image harmonization is a critical task in computer vision, which aims to adjust the foreground to make it compatible with the background. Recent works mainly focus on using global transformations (i.e., normalization and color curve rendering) to achieve visual consistency. However, these models ignore local visual consistency and their huge model sizes limit their harmonization ability on edge devices. In this paper, we propose a hierarchical dynamic network (HDNet) to adapt features from local to global view for better feature transformation in efficient image harmonization. Inspired by the success of various dynamic models, local dynamic (LD) module and mask-aware global dynamic (MGD) module are proposed in this paper. Specifically, LD matches local representations between the foreground and background regions based on semantic similarities, then adaptively adjust every foreground local representation according to the appearance of its $K$-nearest neighbor background regions. In this way, LD can produce more realistic images at a more fine-grained level, and simultaneously enjoy the characteristic of semantic alignment. The MGD effectively applies distinct convolution to the foreground and background, learning the representations of foreground and background regions as well as their correlations to the global harmonization, facilitating local visual consistency for the images much more efficiently. Experimental results demonstrate that the proposed HDNet significantly reduces the total model parameters by more than 80\% compared to previous methods, while still attaining state-of-the-art performance on the popular iHarmony4 dataset. Notably, the HDNet achieves a 4\% improvement in PSNR and a 19\% reduction in MSE compared to the prior state-of-the-art methods.

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Harmonization HAdobe5k(1024$\times$1024) HDNet PSNR 13.24 # 6
MSE 16.69 # 1
fMSE 102.53 # 6
SSIM 0.9931 # 1
Image Harmonization iHarmony4 HDNet MSE 16.55 # 1
PSNR 40.46 # 1
fMSE 179.49 # 12
Image Harmonization iHarmony4 HDNet-lite MSE 24.99 # 8
PSNR 38.63 # 4
fMSE 260.65 # 9