Explaining Object Detectors via Collective Contribution of Pixels

Visual explanations for object detectors are crucial for enhancing their reliability. Since object detectors identify and localize instances by assessing multiple features collectively, generating explanations that capture these collective contributions is critical. However, existing methods focus solely on individual pixel contributions, ignoring the collective contribution of multiple pixels. To address this, we proposed a method for object detectors that considers the collective contribution of multiple pixels. Our approach leverages game-theoretic concepts, specifically Shapley values and interactions, to provide explanations. These explanations cover both bounding box generation and class determination, considering both individual and collective pixel contributions. Extensive quantitative and qualitative experiments demonstrate that the proposed method more accurately identifies important regions in detection results compared to current state-of-the-art methods. The code will be publicly available soon.

Low-Quality Image Detection by Hierarchical VAE

To make an employee roster, photo album, or training dataset of generative models, one needs to collect high-quality images while dismissing low-quality ones. This study addresses a new task of unsupervised detection of low-quality images. We propose a method that not only detects low-quality images with various types of degradation but also provides visual clues of them based on an observation that partial reconstruction by hierarchical variational autoencoders fails for low-quality images. The experiments show that our method outperforms several unsupervised out-of-distribution detection methods and also gives visual clues for low-quality images that help humans recognize them even in thumbnail view.

VarteX: Enhancing Weather Forecast through Distributed Variable Representation

Weather forecasting is essential for various human activities. Recent data-driven models have outperformed numerical weather prediction by utilizing deep learning in forecasting performance. However, challenges remain in efficiently handling multiple meteorological variables. This study proposes a new variable aggregation scheme and an efficient learning framework for that challenge. Experiments show that VarteX outperforms the conventional model in forecast performance, requiring significantly fewer parameters and resources. The effectiveness of learning through multiple aggregations and regional split training is demonstrated, enabling more efficient and accurate deep learning-based weather forecasting.

Matching Non-Identical Objects

Not identical but similar objects are everywhere in the world. Examples include four-legged animals such as dogs and cats, cars of different models, akin flowers in various colors, and countless others. In this study, we address a novel task of matching such non-identical objects. We propose a simple weighting scheme of descriptors that enhances various sparse image matching methods, which were originally designed for matching identical objects captured from different perspectives, and achieve semantically robust matching. The experiments show successful matching between non-identical objects in various cases including domain shift. Further, we present a first evaluation of the robustness of the image matching methods under common corruptions, which is a sort of domain shift, and the proposed method improves the matching in this case as well.

Identifying Important Group of Pixels using Interactions

To better understand the behavior of image classifiers, it is useful to visualize the contribution of individual pixels to the model prediction. In this study, we propose a method, MoXI~($\textbf{Mo}$del e$\textbf{X}$planation by $\textbf{I}$nteractions), that efficiently and accurately identifies a group of pixels with high prediction confidence. The proposed method employs game-theoretic concepts, Shapley values and interactions, taking into account the effects of individual pixels and the cooperative influence of pixels on model confidence. Theoretical analysis and experiments demonstrate that our method better identifies the pixels that are highly contributing to the model outputs than widely-used visualization methods using Grad-CAM, Attention rollout, and Shapley value. While prior studies have suffered from the exponential computational cost in the computation of Shapley value and interactions, we show that this can be reduced to linear cost for our task.

Fourier Analysis on Robustness of Graph Convolutional Neural Networks for Skeleton-based Action Recognition

Using Fourier analysis, we explore the robustness and vulnerability of graph convolutional neural networks (GCNs) for skeleton-based action recognition. We adopt a joint Fourier transform (JFT), a combination of the graph Fourier transform (GFT) and the discrete Fourier transform (DFT), to examine the robustness of adversarially-trained GCNs against adversarial attacks and common corruptions. Experimental results with the NTU RGB+D dataset reveal that adversarial training does not introduce a robustness trade-off between adversarial attacks and low-frequency perturbations, which typically occurs during image classification based on convolutional neural networks. This finding indicates that adversarial training is a practical approach to enhancing robustness against adversarial attacks and common corruptions in skeleton-based action recognition. Furthermore, we find that the Fourier approach cannot explain vulnerability against skeletal part occlusion corruption, which highlights its limitations. These findings extend our understanding of the robustness of GCNs, potentially guiding the development of more robust learning methods for skeleton-based action recognition.

Exploiting Frequency Spectrum of Adversarial Images for General Robustness

In recent years, there has been growing concern over the vulnerability of convolutional neural networks (CNNs) to image perturbations. However, achieving general robustness against different types of perturbations remains challenging, in which enhancing robustness to some perturbations (e.g., adversarial perturbations) may degrade others (e.g., common corruptions). In this paper, we demonstrate that adversarial training with an emphasis on phase components significantly improves model performance on clean, adversarial, and common corruption accuracies. We propose a frequency-based data augmentation method, Adversarial Amplitude Swap, that swaps the amplitude spectrum between clean and adversarial images to generate two novel training images: adversarial amplitude and adversarial phase images. These images act as substitutes for adversarial images and can be implemented in various adversarial training setups. Through extensive experiments, we demonstrate that our method enables the CNNs to gain general robustness against different types of perturbations and results in a uniform performance against all types of common corruptions.

Improving Accuracy of Zero-Shot Action Recognition with Handcrafted Features

With the development of machine learning, datasets for models are getting increasingly larger. This leads to increased data annotation costs and training time, which undoubtedly hinders the development of machine learning. To solve this problem, zero-shot learning is gaining considerable attention. With zero-shot learning, objects can be recognized or classified, even without having been seen before. Nevertheless, the accuracy of this method is still low, thus limiting its practical application. To solve this problem, we propose a video-text matching model, which can learn from handcrafted features. Our model can be used alone to predict the action classes and can also be added to any other model to improve its accuracy. Moreover, our model can be continuously optimized to improve its accuracy. We only need to manually annotate some features, which incurs some labor costs; in many situations, the costs are worth it. The results with UCF101 and HMDB51 show that our model achieves the best accuracy and also improves the accuracies of other models.

Spatiotemporal forecasting of track geometry irregularities with exogenous factors

To ensure the safety of railroad operations, it is important to monitor and forecast track geometry irregularities. A higher safety requires forecasting with a higher spatiotemporal frequency. For forecasting with a high spatiotemporal frequency, it is necessary to capture spatial correlations. Additionally, track geometry irregularities are influenced by multiple exogenous factors. In this study, we propose a method to forecast one type of track geometry irregularity, vertical alignment, by incorporating spatial and exogenous factor calculations. The proposed method embeds exogenous factors and captures spatiotemporal correlations using a convolutional long short-term memory (ConvLSTM). In the experiment, we compared the proposed method with other methods in terms of the forecasting performance. Additionally, we conducted an ablation study on exogenous factors to examine their contribution to the forecasting performance. The results reveal that spatial calculations and maintenance record data improve the forecasting of the vertical alignment.

Game-Theoretic Understanding of Misclassification

This paper analyzes various types of image misclassification from a game-theoretic view. Particularly, we consider the misclassification of clean, adversarial, and corrupted images and characterize it through the distribution of multi-order interactions. We discover that the distribution of multi-order interactions varies across the types of misclassification. For example, misclassified adversarial images have a higher strength of high-order interactions than correctly classified clean images, which indicates that adversarial perturbations create spurious features that arise from complex cooperation between pixels. By contrast, misclassified corrupted images have a lower strength of low-order interactions than correctly classified clean images, which indicates that corruptions break the local cooperation between pixels. We also provide the first analysis of Vision Transformers using interactions. We found that Vision Transformers show a different tendency in the distribution of interactions from that in CNNs, and this implies that they exploit the features that CNNs do not use for the prediction. Our study demonstrates that the recent game-theoretic analysis of deep learning models can be broadened to analyze various malfunctions of deep learning models including Vision Transformers by using the distribution, order, and sign of interactions.