Learning Differential Pyramid Representation for Tone Mapping

2024-12-02

arXiv.org

AI Summary

Existing HDR tone mapping methods rely on hand-crafted Gaussian or Laplacian pyramids, struggling to simultaneously preserve global tone consistency and local detail fidelity—leading to high-frequency loss, halo artifacts, and poor cross-scene generalization. This paper proposes a learnable differential pyramid representation framework that replaces fixed pyramid structures with an end-to-end differentiable architecture. It incorporates a global tone-aware module to ensure holistic naturalness and a local tone refinement module for adaptive contrast enhancement. The framework unifies multi-scale differential features, global contextual modeling, and local contrast control within a single architecture. Evaluated on HDR+ and HDRI Haven datasets, our method achieves PSNR gains of 2.58 dB and 3.31 dB over the second-best approach, respectively. Moreover, it demonstrates superior cross-domain generalization capability for both image and video tone mapping.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Previous tone mapping methods mainly focus on how to enhance tones in low-resolution images and recover details using the high-frequent components extracted from the input image. These methods typically rely on traditional feature pyramids to artificially extract high-frequency components, such as Laplacian and Gaussian pyramids with handcrafted kernels. However, traditional handcrafted features struggle to effectively capture the high-frequency components in HDR images, resulting in excessive smoothing and loss of detail in the output image. To mitigate the above issue, we introduce a learnable Differential Pyramid Representation Network (DPRNet). Based on the learnable differential pyramid, our DPRNet can capture detailed textures and structures, which is crucial for high-quality tone mapping recovery. In addition, to achieve global consistency and local contrast harmonization, we design a global tone perception module and a local tone tuning module that ensure the consistency of global tuning and the accuracy of local tuning, respectively. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art methods, improving PSNR by 2.58 dB in the HDR+ dataset and 3.31 dB in the HDRI Haven dataset respectively compared with the second-best method. Notably, our method exhibits the best generalization ability in the non-homologous image and video tone mapping operation. We provide an anonymous online demo at https://xxxxxx2024.github.io/DPRNet/.

Problem

• Learning adaptive pyramid representation for tone mapping
• Preserving texture fidelity in complex HDR scenes
• Jointly modeling global consistency and local contrast enhancement

Innovation

• Learnable differential pyramid replaces handcrafted pyramid operations
• Global and local tone modules enable perceptual consistency
• Iterative detail enhancement restores full-resolution output progressively