paper-with-me

Papers

Extending Information Bottleneck Attribution to Video Sequences

2025-01-28 · Veronika Solopova, Lucas Schmidt, Dorothea Kolossa

We introduce VIBA, a novel approach for explainable video classification by adapting Information Bottlenecks for Attribution (IBA) to video sequences. While most traditional explainability methods are designed for image models, our IBA framework addresses the need for explainability in temporal models used for video analysis. To demonstrate its effectiveness, we apply VIBA to video deepfake detection, testing it on two architectures: the Xception model for spatial features and a VGG11-based model for capturing motion dynamics through optical flow. Using a custom dataset that reflects recent deepfake generation techniques, we adapt IBA to create relevance and optical flow maps, visually highlighting manipulated regions and motion inconsistencies. Our results show that VIBA generates temporally and spatially consistent explanations, which align closely with human annotations, thus providing interpretability for video classification and particularly for deepfake detection.

📄 PDF Abstract BibTeX arXiv:2501.16889

Code (1)

anonrep/iba-for-video-sequences 공식 구현 pytorch

Tasks

DeepFake DetectionFace SwappingOptical Flow EstimationVideo Classification

Methods 이 논문이 사용한 방법론

Residual Connection 설명 없음
Average Pooling 설명 없음
Depthwise Convolution Depthwise Convolution is a type of convolution where we apply a single convolutional filter for each input channel. In the regular 2D…
Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…
Pointwise Convolution Pointwise Convolution is a type of convolution that uses a 1x1 kernel: a kernel that iterates through every single point. This…
Max Pooling Max Pooling is a pooling operation that calculates the maximum value for patches of a feature map, and uses it to create a downsampled (pooled) feature map. It is usually…
Depthwise Separable Convolution While standard convolution performs the channelwise and spatial-wise computation in one step, Depthwise Separable Convolution
ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification

2025-09-25 · Patrick Knab, Sascha Marton, Philipp J. Schubert, Drago Guggiana 외 arxiv

Concept Bottleneck Models (CBMs) enable interpretable image classification by structuring predictions around human-understandable concepts, but extending this paradigm to video remains challenging due to the difficulty o…

Video ClassificationImage Classification

PCBEAR: Pose Concept Bottleneck for Explainable Action Recognition

2025-04-17 · Jongseo Lee, Wooil Lee, Gyeong-Moon Park, Seong Tae Kim 외

Human action recognition (HAR) has achieved impressive results with deep learning models, but their decision-making process remains opaque due to their black-box nature. Ensuring interpretability is crucial, especially f…

Action RecognitionAction UnderstandingTemporal Action Localization

Comprehensive Information Bottleneck for Unveiling Universal Attribution to Interpret Vision Transformers

2025-07-06 · Jung-Ho Hong, Ho-Joong Kim, Kyu-Sung Jeon, Seong-Whan Lee arxiv

The feature attribution method reveals the contribution of input variables to the decision-making process to provide an attribution map for explanation. Existing methods grounded on the information bottleneck principle c…

Comprehensive Information Bottleneck for Unveiling Universal Attribution to Interpret Vision Transformers

2025-01-01 · CVPR 2025 1 · Jung-Ho Hong, Ho-Joong Kim, Kyu-Sung Jeon, Seong-Whan Lee

The feature attribution method reveals the contribution of input variables to the decision-making process to provide an attribution map for explanation. Existing methods grounded on the information bottleneck princip…

Decision Making

Restricting the Flow: Information Bottlenecks for Attribution

2020-01-02 · ICLR 2020 1 · Karl Schulz, Leon Sixt, Federico Tombari, Tim Landgraf

Attribution methods provide insights into the decision-making of machine learning models like artificial neural networks. For a given input sample, they assign a relevance score to each individual input variable, such as…

Decision Making