This paper presents our Cityscape Aerial image Dataset for Object deTection (CADOT), a new benchmark dataset for object detection, based on raw aerial images sourced from the IGN website and focused on a specific area within the Paris department. CADOT consists of cropped sub-images with irrelevant areas carefully excluded. Due to image complexity, an algorithm with human validation was developed for accurate annotation. CADOT contains 106691 object annotations across 14 categories in COCO format, spanning transportation infrastructure to urban features. We provide distinct splits for training, validation, and testing. The dataset’s complexity is highlighted through comparative performance evaluations against well-established detection models using other existing aerial image datasets such as DIOR. Our results show that CADOT offers a significantly greater challenge than other benchmark aerial image datasets. The CADOT dataset is accessible via CADOT .

Deep neural networks face challenges with distribution shifts across layers, affecting model convergence and performance. While Batch Normalization (BN) addresses these issues, its reliance on a single Gaussian distribution assumption limits adaptability. To overcome this, alternatives like Layer Normalization, Group Normalization, and Mixture Normalization emerged, yet struggle with dynamic activation distributions. We propose ”Context Normalization” (CN), introducing contexts constructed from domain knowledge. CN normalizes data within the same context, enabling local representation. During backpropagation, CN learns normalized parameters and model weights for each context, ensuring efficient convergence and superior performance compared to BN and MN. This approach emphasizes context utilization, offering a fresh perspective on activation normalization in neural networks. We release our code at https://github.com/b-faye/Context-Normalization .

Object detection in Remote Sensing Images (RSI) is a critical task for numerous applications in Earth Observation (EO). Differing from object detection in natural images, object detection in remote sensing images faces challenges of scarcity of annotated data and the presence of small objects represented by only a few pixels. Multi-modal fusion has been determined to enhance the accuracy by fusing data from multiple modalities such as RGB, infrared (IR), lidar, and synthetic aperture radar (SAR). To this end, the fusion of representations at the mid or late stage, produced by parallel subnetworks, is dominant, with the disadvantages of increasing computational complexity in the order of the number of modalities and the creation of additional engineering obstacles. Using the cross-attention mechanism, we propose a novel multi-modal fusion strategy for mapping relationships between different channels at the early stage, enabling the construction of a coherent input by aligning the different modalities. By addressing fusion in the early stage, as opposed to mid or late-stage methods, our method achieves competitive and even superior performance compared to existing techniques. Additionally, we enhance the SWIN transformer by integrating convolution layers into the feed-forward of non-shifting blocks. This augmentation strengthens the model’s capacity to merge separated windows through local attention, thereby improving small object detection. Extensive experiments prove the effectiveness of the proposed multimodal fusion module and the architecture, demonstrating their applicability to object detection in multimodal aerial imagery. Our code is available at here.

Transformers have led to learning-based image compression methods that outperform traditional approaches. However, these methods often suffer from high complexity, limiting their practical application. To address this, various strategies such as knowledge distillation and lightweight architectures have been explored, aiming to enhance efficiency without significantly sacrificing performance. This paper proposes a State Space Model-based Image Compression (SSMIC) architecture. This novel architecture balances performance and computational efficiency, making it suitable for real-world applications. Experimental evaluations confirm the effectiveness of our model in achieving a superior BD-rate while significantly reducing computational complexity and latency compared to competitive learning-based image compression methods.

This paper focuses on the optimization of overparameterized, non-convex low-rank matrix sensing (LRMS)—an essential component in contemporary statistics and machine learning. Recent years have witnessed significant breakthroughs in first-order methods, such as gradient descent, for tackling this non-convex optimization problem. However, the presence of numerous saddle points often prolongs the time required for gradient descent to overcome these obstacles. Moreover, overparameterization can markedly decelerate gradient descent methods, transitioning its convergence rate from linear to sub-linear. In this paper, we introduce an approximated Gauss-Newton (AGN) method for tackling the non-convex LRMS problem. Notably, AGN incurs a computational cost comparable to gradient descent per iteration but converges much faster without being slowed down by saddle points. We prove that, despite the non-convexity of the objective function, AGN achieves Q-linear convergence from random initialization to the global optimal solution. The global Q-linear convergence of AGN represents a substantial enhancement over the convergence of the existing methods for the overparameterized non-convex LRMS. The code for this paper is available at \url{https://github.com/hsijiaxidian/AGN}.

In this paper we propose a new method for document image restoration based on Blind Source Separation. The existing separation methods rely on the general properties of source images such as independence, sparsity, and non-negativity. In this work we show that by exploiting some characteristics of image denoising methods in a play-and-plug scheme, efficient BSS results could be achieved. In particular, we show that the use of BM3D and Non-local Means denoising methods as ingredients in the proposed scheme, which exploits the non-local properties of the image, leads to better image separation in terms of convergence rate and perceptual image quality. We also propose to use the dictionary-learning approach to take the concept of visual chirality into consideration. Finally, we apply the proposed scheme to document image restoration problem and show its advantage through experiments and objective performance evaluation.

In this paper, we propose a new formulation for the blind source separation problem for audio signals with convolutive mixtures to improve the separation performance of Independent Vector Analysis (IVA). The proposed method benefits from both the recently investigated convolutive approximation model and the IVA approaches that take advantages of the cross-band information to avoid permutation alignment. We first exploit the link between the IVA and the Sparse Component Analysis (SCA) methods through the structured sparsity. We then propose a new framework by combining the convolutive narrowband approximation and the Windowed-Group-Lasso (WGL). The optimisation of the model is based on the alternating optimisation approach where the convolutive kernel and the source components are jointly optimised.

We consider the problem of blind source separation for underdetermined convolutive mixtures. Based on the multiplicative narrowband approximation in the time-frequency domain with the help of the short-time-Fourier-transform (STFT) and the sparse representation of the source signals, we formulate the separation problem in an optimization framework. This framework is then generalized based on the recently investigated convolutive narrowband approximation and the statistics of the room impulse response. Algorithms with convergence proof are then employed to solve the proposed optimization problems. The evaluation of the proposed frameworks and algorithms for synthesized and live recorded mixtures are illustrated. The proposed approaches are also tested for mixtures with input noise. Numerical evaluations show the advantages of the proposed methods.

This paper presents a new method for blind source separation problem in reverberant environments with more sources than microphones. Based on the sparsity property in the time-frequency domain and the low-rank assumption of the spectrogram of the source, the STRAUSS (SparsiTy and low-Rank AmplitUde based Source Separation) method is developed. Numerical evaluations show that the proposed method outperforms the existing multichannel NMF approaches, while it is exclusively based on amplitude information.

We consider the problem of extracting the source signals from an under-determined convolutive mixture, assuming known filters. We start from its formulation as a minimization of a convex functional, combining a classical I2discrepancy term between the observed mixture and the one reconstructed from the estimated sources, and a sparse regularization term of source coefficients in a time-frequency domain. We then introduce a first kind of structure, using a hybrid model. Finally, we embed the previously introduced Windowed-Group-Lasso operator into the iterative thresholding/shrinkage algorithm, in order to take into account some structures inside each layers of time-frequency representations. Intensive numerical studies confirm the benefits of such an approach.