Adaptive occlusion sensitivity analysis for visually explaining video recognition networks
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.
Code (1)
Tasks
Decision Makingimage-classificationImage ClassificationSensitivityVideo ClassificationVideo RecognitionSimilar Papers 제목 키워드 기반
Explaining Multimodal Data Fusion: Occlusion Analysis for Wilderness Mapping
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 ObservationSmoothGrad: removing noise by adding noise
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 LearningSensitivityInterpreting Structured Perturbations in Image Protection Methods for Diffusion Models
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
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+1Occlusion Sensitivity Analysis with Augmentation Subspace Perturbation in Deep Feature Space
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