You Only Landmark Once: Lightweight U-Net Face Super Resolution with YOLO-World Landmark Heatmaps

TL;DR

This paper introduces a lightweight face super-resolution method using YOLO-World heatmaps for feature localization, achieving high-quality results without complex architectures or adversarial training.

Contribution

A novel supervision strategy leveraging YOLO-World heatmaps for face super-resolution, eliminating the need for dedicated landmark networks and reducing computational complexity.

Findings

Achieves 8x magnification from 16x16 to 128x128 face images.

Improves reconstruction quality using heatmap-guided loss.

Produces sharper, more realistic face images without adversarial training.

Abstract

Face image super-resolution aims to recover high-resolution facial images from severely degraded inputs. Under extreme upscaling factors, fine facial details are often lost, making accurate reconstruction challenging. Existing methods typically rely on heavy network architectures, adversarial training schemes, or separate alignment networks, increasing model complexity and computational cost. To address these issues, we propose a lightweight U-Net based-architecture designed to reconstructs $128{ \times }128$ facial images from severely degraded $16{ \times }16$ inputs, achieving an $8 \times $ magnification. A key contribution is a novel auxiliary-training-free supervision strategy that leverages heatmaps generated by YOLO-World, an open-vocabulary object detector, to localize key facial features such as eyes, nose, and mouth. These heatmaps are converted into spatial weights to form a heatmap-guided loss that emphasizes reconstruction errors in semantically important regions. Unlike prior methods that require dedicated landmark or alignment networks, our approach directly reuses detector outputs as supervision, maintaining an efficient training and inference pipeline. Experiments on the aligned CelebA dataset demonstrate that the proposed loss consistently improves quantitative metrics and produces sharper, more realistic reconstructions. Overall, our results show that lightweight networks can effectively exploit detection-driven priors for perceptually convincing extreme upscaling, without adversarial training or increased computational cost.