EffiHDR: An Efficient Framework for HDRTV Reconstruction and Enhancement in UHD Systems

Recent advancements in SDRTV-to-HDRTV conversion have yielded impressive results in reconstructing high dynamic range television (HDRTV) videos from standard dynamic range television (SDRTV) videos. However, the practical applications of these techniques are limited for ultra-high definition (UHD) video systems due to their high computational and memory costs. In this paper, we propose EffiHDR, an efficient framework primarily operating in the downsampled space, effectively reducing the computational and memory demands. Our framework comprises a real-time SDRTV-to-HDRTV Reconstruction model and a plug-and-play HDRTV Enhancement model. The SDRTV-to-HDRTV Reconstruction model learns affine transformation coefficients instead of directly predicting output pixels to preserve high-frequency information and mitigate information loss caused by downsampling. It decomposes SDRTV-to-HDR mapping into pixel intensity-dependent and local-dependent affine transformations. The pixel intensity-dependent transformation leverages global contexts and pixel intensity conditions to transform SDRTV pixels to the HDRTV domain. The local-dependent transformation predicts affine coefficients based on local contexts, further enhancing dynamic range, local contrast, and color tone. Additionally, we introduce a plug-and-play HDRTV Enhancement model based on an efficient Transformer-based U-net, which enhances luminance and color details in challenging recovery scenarios. Experimental results demonstrate that our SDRTV-to-HDRTV Reconstruction model achieves real-time 4K conversion with impressive performance. When combined with the HDRTV Enhancement model, our approach outperforms state-of-the-art methods in performance and efficiency.

I. Introduction

High dynamic range television (HDRTV1) videos have significantly improved the visual quality. Compared to standard dynamic range television (SDRTV) videos, HDRTV videos offer a broader range of brightness levels and a wider color gamut, enabling a more realistic representation of the natural world. With the increasing support for HDR displays in devices like televisions, smartphones, and computers, there is a growing demand for converting existing SDRTV videos to HDRTV in order to enhance video broadcasting services. In this paper, we refer to the task of converting SDRTV videos to HDRTV videos as SDRTV-to-HDRTV, adopting the definition provided by HDRTVNet [1]. It is essential to note that the intent of SDRTV-to-HDRTV differs from Inverse Tone Mapping (ITM), as illustrated in Fig. 2. ITM aims to restore the original luminance of the captured scene, focusing on recovering information in saturated areas. In contrast, SDRTV-to-HDRTV aims to improve the visual quality of video on HDR displays, offering comprehensive enhancements in dynamic range, gamut, and local contrast.¹

We add a suffix TV after HDR to indicate content in HDR-TV format and standard(HDR10 and Dolby Vision, etc.).

Fig. 1.

Efficiency comparison between our method and other works. EffiHDR-Base is our SDRTV-to-HDRTV Reconstruction model with the downsampling factor 8. EffiHDR-Enhancer is the cascade of our EffiHDR-Base model and HDRTV Enhancement model.

Fig. 2.

SDRTV-to-HDRTV differs from Inverse Tone Mapping (ITM) functionally. ITM focuses on restoring the original luminance of the captured scene. To display predicted HDR contents, tone mapping may be needed to compress the dynamic range. Because the dynamic range that HDR displays can present is still limited. In contrast, SDRTV-to-HDRTV aims to directly predict the HDRTV videos for HDR display.