SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection

Samuel Wilson, Tobias Fischer, Feras Dayoub, Dimity Miller, Niko Sünderhauf

ICCV 2023 pp. 23565-23576

doi:10.1109/ICCV51070.2023.02154 /iccv/2023/wilson2023iccv-safe/

Abstract

We address the problem of out-of-distribution (OOD) detection for the task of object detection. We show that residual convolutional layers with batch normalisation produce Sensitivity-Aware FEatures (SAFE) that are consistently powerful for distinguishing in-distribution from out-of-distribution detections. We extract SAFE vectors for every detected object, and train a multilayer perceptron on the surrogate task of distinguishing adversarially perturbed from clean in-distribution examples. This circumvents the need for realistic OOD training data, computationally expensive generative models, or retraining of the base object detector. SAFE outperforms the state-of-the-art OOD object detectors on multiple benchmarks by large margins, e.g. reducing the FPR95 by an absolute 30.6% from 48.3% to 17.7% on the OpenImages dataset.

PDF ICCV Semantic Scholar

Cite

Text

Wilson et al. "SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection." International Conference on Computer Vision, 2023. doi:10.1109/ICCV51070.2023.02154

Markdown

[Wilson et al. "SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection." International Conference on Computer Vision, 2023.](https://mlanthology.org/iccv/2023/wilson2023iccv-safe/) doi:10.1109/ICCV51070.2023.02154

BibTeX

@inproceedings{wilson2023iccv-safe,
  title     = {{SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection}},
  author    = {Wilson, Samuel and Fischer, Tobias and Dayoub, Feras and Miller, Dimity and Sünderhauf, Niko},
  booktitle = {International Conference on Computer Vision},
  year      = {2023},
  pages     = {23565-23576},
  doi       = {10.1109/ICCV51070.2023.02154},
  url       = {https://mlanthology.org/iccv/2023/wilson2023iccv-safe/}
}