ATTN-FIQA: Interpretable Attention-based Face Image Quality Assessment with Vision Transformers

• URL: http://arxiv.org/abs/2604.22841v1
• Date: Tue, 21 Apr 2026 12:46:16 GMT
• Title: ATTN-FIQA: Interpretable Attention-based Face Image Quality Assessment with Vision Transformers
• Abstract summary: Face Image Quality Assessment (FIQA) aims to assess the recognition utility of face samples and is essential for reliable face recognition (FR) systems.<n>Recent studies highlighted that these architectures inherently function as saliency learners with attention patterns naturally encoding spatial importance.<n>This work proposes ATTN-FIQA, a novel training-free approach that investigates whether pre-softmax attention scores from pre-trained Vision Transformer-based face recognition models can serve as quality indicators.
• Score: 19.095360516976847
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Face Image Quality Assessment (FIQA) aims to assess the recognition utility of face samples and is essential for reliable face recognition (FR) systems. Existing approaches require computationally expensive procedures such as multiple forward passes, backpropagation, or additional training, and only recent work has focused on the use of Vision Transformers. Recent studies highlighted that these architectures inherently function as saliency learners with attention patterns naturally encoding spatial importance. This work proposes ATTN-FIQA, a novel training-free approach that investigates whether pre-softmax attention scores from pre-trained Vision Transformer-based face recognition models can serve as quality indicators. We hypothesize that attention magnitudes intrinsically encode quality: high-quality images with discriminative facial features enable strong query-key alignments producing focused, high-magnitude attention patterns, while degraded images generate diffuse, low-magnitude patterns. ATTN-FIQA extracts pre-softmax attention matrices from the final transformer block, aggregate multi-head attention information across all patches, and compute image-level quality scores through simple averaging, requiring only a single forward pass through pre-trained models without architectural modifications, backpropagation, or additional training. Through comprehensive evaluation across eight benchmark datasets and four FR models, this work demonstrates that attention-based quality scores effectively correlate with face image quality and provide spatial interpretability, revealing which facial regions contribute most to quality determination.

Related papers

• Surveillance Facial Image Quality Assessment: A Multi-dimensional Dataset and Lightweight Model [59.39390911456143]
  We propose the first comprehensive study on surveillance facial image quality assessment (SFIQA)<n>SFIQA-Bench consists of 5,004 surveillance facial images captured by three widely deployed surveillance cameras in real-world scenarios.<n>A subjective experiment is conducted to collect six dimensional quality ratings, including noise, sharpness, colorfulness, contrast, fidelity and overall quality.
  arXiv  Detail & Related papers  (2026-02-07T06:51:03Z)
• IQPFR: An Image Quality Prior for Blind Face Restoration and Beyond [56.99331967165238]
  Blind Face Restoration (BFR) addresses the challenge of reconstructing degraded low-quality (LQ) facial images into high-quality (HQ) outputs.<n>We propose a novel framework that incorporates an Image Quality Prior (IQP) derived from No-Reference Image Quality Assessment (NR-IQA) models.<n>Our method outperforms state-of-the-art techniques across multiple benchmarks.
  arXiv  Detail & Related papers  (2025-03-12T11:39:51Z)
• DSL-FIQA: Assessing Facial Image Quality via Dual-Set Degradation Learning and Landmark-Guided Transformer [23.70791030264281]
  Generic Face Image Quality Assessment (GFIQA) evaluates the perceptual quality of facial images. We present a novel transformer-based method for GFIQA, which is aided by two unique mechanisms.
  arXiv  Detail & Related papers  (2024-06-13T23:11:25Z)
• Rank-based No-reference Quality Assessment for Face Swapping [88.53827937914038]
  The metric of measuring the quality in most face swapping methods relies on several distances between the manipulated images and the source image. We present a novel no-reference image quality assessment (NR-IQA) method specifically designed for face swapping.
  arXiv  Detail & Related papers  (2024-06-04T01:36:29Z)
• DP-IQA: Utilizing Diffusion Prior for Blind Image Quality Assessment in the Wild [73.6767681305851]
  Blind image quality assessment (IQA) in the wild presents significant challenges.<n>Given the difficulty in collecting large-scale training data, leveraging limited data to develop a model with strong generalization remains an open problem.<n>Motivated by the robust image perception capabilities of pre-trained text-to-image (T2I) diffusion models, we propose a novel IQA method, diffusion priors-based IQA.
  arXiv  Detail & Related papers  (2024-05-30T12:32:35Z)
• Assessing Image Quality Using a Simple Generative Representation [34.173947968362675]
  VAE-QA is a simple and efficient method for predicting image quality in the presence of a full-reference. We evaluate our approach on four standard benchmarks and find that it significantly improves generalization across datasets.
  arXiv  Detail & Related papers  (2024-04-28T13:18:47Z)
• Multi-Modal Prompt Learning on Blind Image Quality Assessment [65.0676908930946]
  Image Quality Assessment (IQA) models benefit significantly from semantic information, which allows them to treat different types of objects distinctly. Traditional methods, hindered by a lack of sufficiently annotated data, have employed the CLIP image-text pretraining model as their backbone to gain semantic awareness. Recent approaches have attempted to address this mismatch using prompt technology, but these solutions have shortcomings. This paper introduces an innovative multi-modal prompt-based methodology for IQA.
  arXiv  Detail & Related papers  (2024-04-23T11:45:32Z)
• Blind Image Quality Assessment via Transformer Predicted Error Map and Perceptual Quality Token [19.67014524146261]
  No-reference image quality assessment (NR-IQA) has gained increasing attention recently. We propose a Transformer based NR-IQA model using a predicted objective error map and perceptual quality token. Our proposed method outperforms the current state-of-the-art in both authentic and synthetic image databases.
  arXiv  Detail & Related papers  (2023-05-16T11:17:54Z)
• CR-FIQA: Face Image Quality Assessment by Learning Sample Relative Classifiability [2.3624125155742055]
  We propose a novel learning paradigm that learns internal network observations during the training process. Our proposed CR-FIQA uses this paradigm to estimate the face image quality of a sample by predicting its relative classifiability. We demonstrate the superiority of our proposed CR-FIQA over state-of-the-art (SOTA) FIQA algorithms.
  arXiv  Detail & Related papers  (2021-12-13T12:18:43Z)
• Inducing Predictive Uncertainty Estimation for Face Recognition [102.58180557181643]
  We propose a method for generating image quality training data automatically from'mated-pairs' of face images. We use the generated data to train a lightweight Predictive Confidence Network, termed as PCNet, for estimating the confidence score of a face image.
  arXiv  Detail & Related papers  (2020-09-01T17:52:00Z)
• Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild [98.48284827503409]
  We develop a textitunified BIQA model and an approach of training it for both synthetic and realistic distortions. We employ the fidelity loss to optimize a deep neural network for BIQA over a large number of such image pairs. Experiments on six IQA databases show the promise of the learned method in blindly assessing image quality in the laboratory and wild.
  arXiv  Detail & Related papers  (2020-05-28T13:35:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.