# Content-Variant Reference Image Quality Assessment via Knowledge Distillation arXiv: https://arxiv.org/abs/2202.13123 ยท HTML: https://mlsanity.com/paper/2202.13123/ ## Key facts - Cited by 14 papers in the map (data through 14 Oct 2026). - Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation. - Builds on 18 of its 41 references that are in the map. - First posted on arXiv 26 Feb 2022. ## Placed in - Knowledge distillation / [General knowledge distillation methods](https://mlsanity.com/distill/general-knowledge-distillation-methods/) ## Builds on (most cited first, 18 of 18) - [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017) - [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015) - [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018) - [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015) - [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021) - [Image Super-Resolution Using Deep Convolutional Networks](https://mlsanity.com/paper/1501.00092/) (2014) - [Do Deep Nets Really Need to be Deep?](https://mlsanity.com/paper/1312.6184/) (2013) - [KonIQ-10k: An ecologically valid database for deep learning of blind image quality assessment](https://mlsanity.com/paper/1910.06180/) (2019) - [Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment](https://mlsanity.com/paper/1612.01697/) (2016) - [Knowledge Distillation by On-the-Fly Native Ensemble](https://mlsanity.com/paper/1806.04606/) (2018) - [Cross Modal Distillation for Supervision Transfer](https://mlsanity.com/paper/1507.00448/) (2015) - [PieAPP: Perceptual Image-Error Assessment through Pairwise Preference](https://mlsanity.com/paper/1806.02067/) (2018) - [Modality Distillation with Multiple Stream Networks for Action Recognition](https://mlsanity.com/paper/1806.07110/) (2018) - [Transformer for Image Quality Assessment](https://mlsanity.com/paper/2101.01097/) (2020) - [dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs](https://mlsanity.com/paper/1904.06505/) (2019) - [Norm-in-Norm Loss with Faster Convergence and Better Performance for Image Quality Assessment](https://mlsanity.com/paper/2008.03889/) (2020) - [Perceptual Image Quality Assessment with Transformers](https://mlsanity.com/paper/2104.14730/) (2021) - [Robust Re-Identification by Multiple Views Knowledge Distillation](https://mlsanity.com/paper/2007.04174/) (2020) ## Built on by (most cited first, 14 of 14) - [Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution](https://mlsanity.com/paper/2501.11561/) (2025) - [Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Models](https://mlsanity.com/paper/2312.08962/) (2023) - [Enhancing Descriptive Image Quality Assessment with A Large-scale Multi-modal Dataset](https://mlsanity.com/paper/2405.18842/) (2024) - [LMM4Edit: Benchmarking and Evaluating Multimodal Image Editing with LMMs](https://mlsanity.com/paper/2507.16193/) (2025) - [EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing](https://mlsanity.com/paper/2603.14916/) (2026) - [BPCLIP: A Bottom-up Image Quality Assessment from Distortion to Semantics Based on CLIP](https://mlsanity.com/paper/2506.17969/) (2025) - [How far have we gone in Generative Image Restoration? A study on its capability, limitations and evaluation practices](https://mlsanity.com/paper/2603.05010/) (2026) - [SAM-IQA: Can Segment Anything Boost Image Quality Assessment?](https://mlsanity.com/paper/2307.04455/) (2023) - [Non-Aligned Reference Image Quality Assessment for Novel View Synthesis](https://mlsanity.com/paper/2511.08155/) (2025) - [Fine-Grained Human Pose Editing Assessment via Layer-Selective MLLMs](https://mlsanity.com/paper/2601.10369/) (2026) - [EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement](https://mlsanity.com/paper/2605.07457/) (2026) - [Development of Real-time Rendering Technology for High-Precision Models in Autonomous Driving](https://mlsanity.com/paper/2302.00291/) (2023) - [Less is More: Learning Reference Knowledge Using No-Reference Image Quality Assessment](https://mlsanity.com/paper/2312.00591/) (2023) - [BlindPSNR: A No-Reference Fidelity Predictor for Low-Light Image Enhancement](https://mlsanity.com/paper/2607.27628/) (2026) ## Abstract (arXiv) Generally, humans are more skilled at perceiving differences between high-quality (HQ) and low-quality (LQ) images than directly judging the quality of a single LQ image. This situation also applies to image quality assessment (IQA). Although recent no-reference (NR-IQA) methods have made great progress to predict image quality free from the reference image, they still have the potential to achieve better performance since HQ image information is not fully exploited. In contrast, full-reference (FR-IQA) methods tend to provide more reliable quality evaluation, but its practicability is affected by the requirement for pixel-level aligned reference images. To address this, we firstly propose the content-variant reference method via knowledge distillation (CVRKD-IQA). Specifically, we use non-aligned reference (NAR) images to introduce various prior distributions of high-quality images. The comparisons of distribution differences between HQ and LQ images can help our model better assess the image quality. Further, the knowledge distillation transfers more HQ-LQ distribution difference information from the FR-teacher to the NAR-student and stabilizing CVRKD-IQA performance. Moreover, to fully mine the local-global combined information, while achieving faster inference speed, our model directly processes multiple image patches from the input with the MLP-mixer. Cross-dataset experiments verify that our model can outperform all NAR/NR-IQA SOTAs, even reach comparable performance with FR-IQA methods on some occasions. Since the content-variant and non-aligned reference HQ images are easy to obtain, our model can support more IQA applications with its relative robustness to content variations. Our code and more detailed elaborations of supplements are available: https://github.com/guanghaoyin/CVRKD-IQA.