Problem
Research questions and friction points this paper is trying to address.
UAV infrared image super-resolution
multi-loss training instability
gradient conflict
frequency-domain objectives
low contrast
OGG-FR: Orthogonal Gradient Gaming and Frequency Rectification for Unmanned Aerial Vehicle Infrared Image Super-Resolution
🤖 AI Summary
This work addresses the challenges of unstable multi-loss training and conflicting gradients between pixel and frequency domains in infrared unmanned aerial vehicle (UAV) image super-resolution. To this end, the authors propose a plug-and-play optimization framework that decomposes frequency-domain gradients into parallel redundant and orthogonal innovative components. By integrating a conflict-aware mechanism to dynamically adjust update strategies, and leveraging orthogonal gradient gaming together with high-frequency residual confidence-guided frequency correction, the method enhances both training stability and reconstruction quality. Built upon techniques including Multiple Gradient Descent Algorithm (MGDA), gradient orthogonal decomposition, and variance correction, the proposed approach achieves significant performance gains on UAV thermal imaging benchmarks under ×4 and ×8 super-resolution settings for both bicubic (BI) and blur-downscale (BD) degradations, with gradient analysis confirming its efficacy.
📝 Abstract
Unmanned aerial vehicle (UAV) infrared image super-resolution aims to recover weak thermal structures for deployment on resource-constrained platforms; lightweight models are therefore preferred, but multi-loss training can be unstable. A common strategy combines pixel-domain and frequency-domain objectives; however, low contrast, limited high-frequency content, and sensor-specific noise often make their gradients weakly aligned or conflicting. To address this optimization ambiguity, we propose Orthogonal Gradient Gaming and Frequency Rectification (OGG-FR), a plug-and-play optimization framework that decomposes the frequency gradient into a redundant parallel component and an orthogonal innovation component relative to the pixel gradient. In the conflict regime, OGG-FR computes a safe base gradient using the Multiple Gradient Descent Algorithm (MGDA) and adds a variance-rectified orthogonal innovation; in the compatible regime, it discards redundant parallel information and injects the orthogonal innovation according to a confidence score estimated from the high-frequency residual. Experimental results on the UAV thermal benchmark show broad gains under BI and BD degradations at $\times 4$ and $\times 8$ scales, while gradient analyses support the effectiveness of the proposed conflict-aware update rule.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Orthogonal Gradient Decomposition
Frequency Rectification
Multi-loss Optimization
Infrared Image Super-Resolution
Gradient Conflict Resolution
🔎 Similar Papers
No similar papers found.
