Client contribution evaluation is crucial in federated learning(FL) to effectively select influential clients. Contrary to data valuation in centralized settings, client contribution evaluation in FL faces a lack of data accessibility and consequently challenges stable quantification of the impact of data heterogeneity. To address this instability of client contribution evaluation, we introduce an empirical method, Federated Client Contribution Evaluation through Accuracy Approximation(FedCCEA), which exploits data size as a tool for client contribution evaluation. After several FL simulations, FedCCEA approximates the test accuracy using the sampled data size and extracts the client contribution from the trained accuracy approximator. In addition, FedCCEA grants data size diversification , which reduces the massive variation in accuracy resulting from game-theoretic strategies. Several experiments have shown that FedCCEA strengthens the robustness to diverse heterogeneous data environments and the practicality of partial participation.

Face recognition, including emotion classification and face attribute classification, has seen tremendous progress during the last decade owing to the use of deep learning. Large-scale data collected from numerous users have been the driving force in this growth. However, face images containing the identities of the owner can potentially cause severe privacy leakage if linked to other sensitive biometric information. The novel discrete cosine transform (DCT) coefficient cutting method (DCC) proposed in this study combines DCT and pixelization to protect the privacy of the image. However, privacy is subjective, and it is not guaranteed that the transformed image will preserve privacy. To overcome this, a user study was conducted on whether DCC really preserves privacy. To this end, convolutional neural networks were trained for face recognition and face attribute classification tasks. Our survey and experiments demonstrate that a face recognition deep learning model can be trained with images that most people think preserve privacy at a manageable cost in classification accuracy.

Traditional tabular classifiers provide explainable decision-making with interpretable features (concepts). However, using their explainability in vision tasks has been limited due to the pixel representation of images. In this paper, we design Img2Tabs that classify images by concepts to harness the explainability of tabular classifiers. Img2Tabs encode image pixels into tabular features by StyleGAN inversion. Since not all of the resulting features are class-relevant or interpretable due to their generative nature, the Img2Tab classifier should automatically discover class-relevant concepts from the StyleGAN features. Thus, we propose a novel algorithm using the Wasserstein-1 metric to quantify class-relevancy and interpretability simultaneously. By this method of concept visualization, we quantitatively investigate whether important features extracted by tabular classifiers are class-relevant concepts. Consequently, we determine the most effective classifier for Img2Tabs in terms of discovering class-relevant concepts automatically from StyleGAN features. In evaluations, we demonstrate concept-based explanations through importance and visualization. Img2Tab achieves top-1 accuracy on par with CNN classifiers and deep feature learning baselines. Additionally, we show that users can interactively debug Img2Tab classifier to prevent erroneous decision-making from data bias without sacrificing accuracy. The source and demo code for Img2Tab are available at https://github.com/songsnim/Img2Tab_pytorch

Non-line-of-sight (NLOS) imaging is aimed at visualizing hidden scenes from an observer’s (e.g., camera) viewpoint. Typically, hidden scenes are reconstructed using diffused signals that emit light sources using optical equipment and are reflected multiple times. Optical systems are commonly adopted in NLOS imaging because lasers can transport energy and focus light over long distances without loss. In contrast, we propose NLOS imaging using acoustic equipment inspired by echolocation. Existing acoustic NLOS is a computational method motivated by seismic imaging that analyzes the geometry of underground structures. However, this physical method is susceptible to noise and requires a clear signal, resulting in long data acquisition times. Therefore, we reduced the scan time by modifying the echoes to be collected simultaneously rather than sequentially. Then, we propose end-to-end deep-learning models to overcome the challenges of echoes interfering with each other. We designed three distinctive architectures: an encoder that extracts features by dividing multi-channel echoes into groups and merging them hierarchically, a generator that constructs an image of the hidden object, and a discriminator that compares the generated image with the ground-truth image. The proposed model successfully reconstructed the outline of the hidden objects.

Acoustic non-line-of-sight (NLOS) imaging aims to reconstruct hidden scenes by analyzing reflections of acoustic waves. Despite recent developments in the field, existing methods still have limitations such as sensitivity to noise in a physical model and difficulty in reconstructing unseen objects in a deep learning model. To address these limitations, we propose a novel cross-modal knowledge distillation (CMKD) approach for acoustic NLOS imaging. Our method transfers knowledge from a well-trained image network to an audio network, effectively combining the strengths of both modalities. As a result, it is robust to noise and superior in reconstructing unseen objects. Additionally, we evaluate real-world datasets and demonstrate that the proposed method outperforms state-of-the-art methods in acoustic NLOS imaging. The experimental results indicate that CMKD is an effective solution for addressing the limitations of current acoustic NLOS imaging methods. Our code, model, and data are available at https://github.com/shineh96/Acoustic-NLOS-CMKD.