mlsanityData to 14 Oct 2026

Benchmarking CLIP for Zero-Shot Face and Periocular Gender Estimation

arXiv 2610.06102 ↗First posted 5 Oct 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

We investigate CLIP for zero-shot gender estimation from full-face and periocular images. Three CLIP backbones are evaluated on 11,299 frontal images from Adience using image-text similarity with male/female prompts, achieving 95.54% full-face accuracy without task-specific training. For periocular, zero-shot predictions are strongly biased towards males, primarily due to a misaligned decision boundary. Threshold alignment substantially reduces this bias, reaching 85.29% accuracy. Linear SVMs trained on CLIP features provide only marginal gains, with a best periocular accuracy of 86.17%, approximately 2.8% above previous Adience results in the literature. Nevertheless, the gap with full-face performance confirms the greater difficulty of periocular gender estimation

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 19 of its 30 references that are in the map. First posted on arXiv 5 Oct 2026.

Builds on

19 of 30 references.

References · 19

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2016Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networkscited by 903
  3. 2023MiVOLO: Multi-input Transformer for Age and Gender Estimationcited by 42
  4. 2024How Good is ChatGPT at Face Biometrics? A First Look into Recognition, Soft Biometrics, and Explainabilitycited by 34
  5. 2025FaceXBench: Evaluating Multimodal LLMs on Face Understandingcited by 24
  6. 2024ChatGPT and biometrics: an assessment of face recognition, gender detection, and age estimation capabilitiescited by 21
  7. 202550 Years of Automated Face Recognitioncited by 15
  8. 2025Benchmarking Foundation Models for Zero-Shot Biometric Taskscited by 14
  9. 2024Face-MLLM: A Large Face Perception Modelcited by 13
  10. 2022Periocular Biometrics and its Relevance to Partially Masked Faces: A Surveycited by 10
  11. 2022Periocular Biometrics: A Modality for Unconstrained Scenarioscited by 10
  12. 2021Facial Masks and Soft-Biometrics: Leveraging Face Recognition CNNs for Age and Gender Prediction on Mobile Ocular Imagescited by 8
7 more
  1. 2024AdaDistill: Adaptive Knowledge Distillation for Deep Face Recognitioncited by 8
  2. 2024Model Compression Techniques in Biometrics Applications: A Surveycited by 5
  3. 2024Combined CNN and ViT features off-the-shelf: Another astounding baseline for recognitioncited by 4
  4. 2017Recognizing Gender from Human Facial Regions using Genetic Algorithmcited by 3
  5. 2024Deep Network Pruning: A Comparative Study on CNNs in Face Recognitioncited by 3
  6. 2025FaceInsight: A Multimodal Large Language Model for Face Perceptioncited by 3
  7. 2024A Hybrid Transformer-Sequencer approach for Age and Gender classification from in-wild facial imagescited by 1

Built on by

Not cited yet.