Regmi, S;
Panthi, B;
Dotel, S;
Gyawali, PK;
Stoyanov, D;
Bhattarai, B;
(2024)
T2FNorm: Train-time Feature Normalization for OOD Detection in Image Classification.
In:
IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.
(pp. pp. 153-162).
IEEE: Seattle, WA, USA.
Preview |
Text
Regmi_T2FNorm_Train-time_Feature_Normalization_for_OOD_Detection_in_Image_Classification_CVPRW_2024_paper.pdf - Accepted Version Download (5MB) | Preview |
Abstract
Neural networks are notorious for being overconfident predictors, posing a significant challenge to their safe deployment in real-world applications. While feature normalization has garnered considerable attention within the deep learning literature, current train-time regularization methods for Out-of-Distribution(OOD) detection are yet to fully exploit this potential. Indeed, the naive incorporation of feature normalization within neural networks does not guarantee substantial improvement in OOD detection performance. In this work, we introduce T2FNorm, a novel approach to transforming features to hyperspherical space during training, while employing non-transformed space for OOD-scoring purposes. This method yields a surprising enhancement in OOD detection capabilities without compromising model accuracy in in-distribution(ID). Our investigation demonstrates that the proposed technique substantially diminishes the norm of the features of all samples, more so in the case of out-of-distribution samples, thereby addressing the prevalent concern of overconfidence in neural networks. The proposed method also significantly improves various post-hoc OOD detection methods.
| Type: | Proceedings paper |
|---|---|
| Title: | T2FNorm: Train-time Feature Normalization for OOD Detection in Image Classification |
| Event: | 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) |
| Dates: | 17 Jun 2024 - 18 Jun 2024 |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1109/CVPRW63382.2024.00020 |
| Publisher version: | https://doi.org/10.1109/CVPRW63382.2024.00020 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Training, Measurement, Deep learning, Computer vision, Accuracy, Conferences, Neural networks |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10198868 |
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
