Efficient, Validation-Free Intrinsic Quality Estimation for Large-Scale Face Recognition Datasets

Zhichao Chen, Yongle Zhao, Kaicheng Yang, Meng Yang, Yin Xie, Ziyong Feng
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:18557-18567, 2026.

Abstract

We propose Intrinsic Quality (IQ), a validation-free metric designed to estimate the inherent potential of face recognition (FR) datasets to produce high-performance models without the need for full-scale training. IQ integrates two components: (i) a Neighbor-Consistency Score that quantifies local identity label agreement via nearest neighbors, and (ii) Global Representation Subspace Complexity (Effective Rank, ER), which captures the underlying embedding geometry and dataset diversity. IQ allows for rapid evaluation using lightweight proxy models or data subsets, facilitating dataset diagnosis and curation prior to resource-intensive full-scale training. We describe an experimental protocol tailored to clean, noisy, and mixed-quality FR datasets, and outline evaluation methodologies to validate IQ’s predictive power for downstream performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26gw, title = {Efficient, Validation-Free Intrinsic Quality Estimation for Large-Scale Face Recognition Datasets}, author = {Chen, Zhichao and Zhao, Yongle and Yang, Kaicheng and Yang, Meng and Xie, Yin and Feng, Ziyong}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {18557--18567}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/chen26gw/chen26gw.pdf}, url = {https://proceedings.mlr.press/v306/chen26gw.html}, abstract = {We propose Intrinsic Quality (IQ), a validation-free metric designed to estimate the inherent potential of face recognition (FR) datasets to produce high-performance models without the need for full-scale training. IQ integrates two components: (i) a Neighbor-Consistency Score that quantifies local identity label agreement via nearest neighbors, and (ii) Global Representation Subspace Complexity (Effective Rank, ER), which captures the underlying embedding geometry and dataset diversity. IQ allows for rapid evaluation using lightweight proxy models or data subsets, facilitating dataset diagnosis and curation prior to resource-intensive full-scale training. We describe an experimental protocol tailored to clean, noisy, and mixed-quality FR datasets, and outline evaluation methodologies to validate IQ’s predictive power for downstream performance.} }
Endnote
%0 Conference Paper %T Efficient, Validation-Free Intrinsic Quality Estimation for Large-Scale Face Recognition Datasets %A Zhichao Chen %A Yongle Zhao %A Kaicheng Yang %A Meng Yang %A Yin Xie %A Ziyong Feng %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-chen26gw %I PMLR %P 18557--18567 %U https://proceedings.mlr.press/v306/chen26gw.html %V 306 %X We propose Intrinsic Quality (IQ), a validation-free metric designed to estimate the inherent potential of face recognition (FR) datasets to produce high-performance models without the need for full-scale training. IQ integrates two components: (i) a Neighbor-Consistency Score that quantifies local identity label agreement via nearest neighbors, and (ii) Global Representation Subspace Complexity (Effective Rank, ER), which captures the underlying embedding geometry and dataset diversity. IQ allows for rapid evaluation using lightweight proxy models or data subsets, facilitating dataset diagnosis and curation prior to resource-intensive full-scale training. We describe an experimental protocol tailored to clean, noisy, and mixed-quality FR datasets, and outline evaluation methodologies to validate IQ’s predictive power for downstream performance.
APA
Chen, Z., Zhao, Y., Yang, K., Yang, M., Xie, Y. & Feng, Z.. (2026). Efficient, Validation-Free Intrinsic Quality Estimation for Large-Scale Face Recognition Datasets. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:18557-18567 Available from https://proceedings.mlr.press/v306/chen26gw.html.

Related Material