paper-with-me

Papers

Adaptive occlusion sensitivity analysis for visually explaining video recognition networks

2022-07-26 · Tomoki Uchiyama, Naoya Sogi, Satoshi Iizuka, Koichiro Niinuma, Kazuhiro Fukui

This paper proposes a method for visually explaining the decision-making process of video recognition networks with a temporal extension of occlusion sensitivity analysis, called Adaptive Occlusion Sensitivity Analysis (AOSA). The key idea here is to occlude a specific volume of data by a 3D mask in an input 3D temporal-spatial data space and then measure the change degree in the output score. The occluded volume data that produces a larger change degree is regarded as a more critical element for classification. However, while the occlusion sensitivity analysis is commonly used to analyze single image classification, applying this idea to video classification is not so straightforward as a simple fixed cuboid cannot deal with complicated motions. To solve this issue, we adaptively set the shape of a 3D occlusion mask while referring to motions. Our flexible mask adaptation is performed by considering the temporal continuity and spatial co-occurrence of the optical flows extracted from the input video data. We further propose a novel method to reduce the computational cost of the proposed method with the first-order approximation of the output score with respect to an input video. We demonstrate the effectiveness of our method through various and extensive comparisons with the conventional methods in terms of the deletion/insertion metric and the pointing metric on the UCF101 dataset and the Kinetics-400 and 700 datasets.

📄 PDF Abstract BibTeX arXiv:2207.12859

Code (1)

uchiyama33/aosa 공식 구현 pytorch

Tasks

Decision Makingimage-classificationImage ClassificationSensitivityVideo ClassificationVideo Recognition

Similar Papers 제목 키워드 기반

Explaining Multimodal Data Fusion: Occlusion Analysis for Wilderness Mapping

2023-04-05 · Burak Ekim, Michael Schmitt

Jointly harnessing complementary features of multi-modal input data in a common latent space has been found to be beneficial long ago. However, the influence of each modality on the models decision remains a puzzle. This…

Earth Observation

SmoothGrad: removing noise by adding noise

2017-06-12 · Daniel Smilkov, Nikhil Thorat, Been Kim, Fernanda Viégas 외

Explaining the output of a deep network remains a challenge. In the case of an image classifier, one type of explanation is to identify pixels that strongly influence the final decision. A starting point for this strateg…

Interpretable Machine LearningSensitivity

Interpreting Structured Perturbations in Image Protection Methods for Diffusion Models

2025-12-09 · Michael R. Martin, Garrick Chan, Kwan-Liu Ma arxiv

Recent image protection mechanisms such as Glaze and Nightshade introduce imperceptible, adversarially designed perturbations intended to disrupt downstream text-to-image generative models. While their empirical effectiv…

Explaining Time Series by Counterfactuals

2019-09-25 · Sana Tonekaboni, Shalmali Joshi, David Duvenaud, Anna Goldenberg

We propose a method to automatically compute the importance of features at every observation in time series, by simulating counterfactual trajectories given previous observations. We define the importance of each observa…

counterfactualFeature ImportanceSensitivityTime Series+1

Occlusion Sensitivity Analysis with Augmentation Subspace Perturbation in Deep Feature Space

2023-11-25 · Pedro Valois, Koichiro Niinuma, Kazuhiro Fukui

Deep Learning of neural networks has gained prominence in multiple life-critical applications like medical diagnoses and autonomous vehicle accident investigations. However, concerns about model transparency and biases p…

Sensitivity