Dense image prediction tasks demand features with strong category information and precise spatial boundary details at high resolution. To achieve this, modern hierarchical models often utilize feature fusion, directly adding upsampled coarse features from deep layers and high-resolution features from lower levels. In this paper, we observe rapid variations in fused feature values within objects, resulting in intra-category inconsistency due to disturbed high-frequency features. Additionally, blurred boundaries in fused features lack accurate high frequency, leading to boundary displacement. Building upon these observations, we propose Frequency-Aware Feature Fusion (FreqFusion), integrating an Adaptive Low-Pass Filter (ALPF) generator, an offset generator, and an Adaptive High-Pass Filter (AHPF) generator. The ALPF generator predicts spatially-variant low-pass filters to attenuate high-frequency components within objects, reducing intra-class inconsistency during upsampling. The offset generator refines large inconsistent features and thin boundaries by replacing inconsistent features with more consistent ones through resampling, while the AHPF generator enhances high-frequency detailed boundary information lost during downsampling. Comprehensive visualization and quantitative analysis demonstrate that FreqFusion effectively improves feature consistency and sharpens object boundaries. Extensive experiments across various dense prediction tasks confirm its effectiveness.

High spatial frequency information, including fine details like textures, significantly contributes to the accuracy of semantic segmentation. However, according to the Nyquist-Shannon Sampling Theorem, high-frequency components are vulnerable to aliasing or distortion when propagating through downsampling layers such as strided-convolution. Here, we propose a novel Spatial Frequency Modulation (SFM) that modulates high-frequency features to a lower frequency before downsampling and then demodulates them back during upsampling. Specifically, we implement modulation through adaptive resampling (ARS) and design a lightweight add-on that can densely sample the high-frequency areas to scale up the signal, thereby lowering its frequency in accordance with the Frequency Scaling Property. We also propose Multi-Scale Adaptive Upsampling (MSAU) to demodulate the modulated feature and recover high-frequency information through non-uniform upsampling This module further improves segmentation by explicitly exploiting information interaction between densely and sparsely resampled areas at multiple scales. Both modules can seamlessly integrate with various architectures, extending from convolutional neural networks to transformers. Feature visualization and analysis demonstrate that our method effectively alleviates aliasing while successfully retaining details after demodulation. As a result, the proposed approach considerably enhances existing state-of-the-art segmentation models (e.g., Mask2Former-Swin-T +1.5 mIoU, InternImage-T +1.4 mIoU on ADE20 K). Furthermore, ARS also enhances the performance of powerful Deformable Convolution (+0.8 mIoU on Cityscapes) by maintaining relative positional order during non-uniform sampling. Finally, we validate the broad applicability and effectiveness of SFM by extending it to image classification, adversarial robustness, instance segmentation, and panoptic segmentation tasks.

Instance segmentation in aerial images is of great significance for remote sensing applications, and it is inherently more challenging because of cluttered background, extremely dense and small objects, and objects with arbitrary orientations. Besides, current mainstream CNN-based methods often suffer from the trade-off between labeling cost and performance. To address these problems, we present a pipeline of hybrid supervision. In the pipeline, we design an ancillary segmentation model with the bounding box attention module and bounding box filter module. It is able to generate accurate pseudo pixel-wise labels from real-world aerial images for training any instance segmentation models. Specifically, bounding box attention module can effectively suppress the noise in cluttered background and improve the capability of segmenting small objects. Bounding box filter module works as a filter which removes the false positives caused by cluttered background and densely distributed objects. Our ancillary segmentation model can locate object pixel-wisely instead of relying on horizontal bounding box prediction, which has better adaptability to arbitrary oriented objects. Furthermore, oriented bounding box labels are utilized for handling arbitrary oriented objects. Experiments on iSAID dataset show that the proposed method can achieve comparable performance (32.1 $AP$) to fully supervised methods (33.9 $AP$), which is obviously higher than weakly supervised setting (26.5 $AP$), when using only 10% pixel-wise labels.

The multilevel tiled map service is widely used and serves as a kind of digital infrastructure. These map tiles are usually rendered from vector data, whose update needs to walk or drive with professional equipment to check every point of interest. This leads to inconvenience and expensive cost in timely updating maps. Compared with vector data, aerial images are much easier and cheaper to obtain. In this article, we propose a novel multilevel map (MLM) generation framework that can automatically generate accurate and consistent maps with multiple zoom levels from aerial images. It consists of a level-aware map generator and a consistency-aware map generator. The level-aware map generator is able to generate accurate initial maps with realistic details for each zoom level. The consistency-aware map generator regards the initial maps at each zoom level as a sequence and builds the connection between them, so as to guarantee content consistency between maps at different zoom levels. Furthermore, we collect a large-scale high-quality dataset called MLM for map generation at multiple zoom levels. Experiments on our MLM dataset show that our method outperforms the previous state-of-the-art map generation methods on both comprehensive quantitative metrics and perceptual quality.

With the rapid development of remote sensing technology, the quality of satellite imagery is getting higher, which contains rich cartographic information that can be translated into maps. However, existing methods either only focus on generating single-level map, or do not fully consider the challenges of multi-level translation from satellite imageries, i.e ., the large domain gap, level-dependent content differences, and main content consistency. In this paper, we propose a novel level-aware fusion network for the satellite imagery based multi-level map generation task. It aims to tackle these three challenges. To deal with the large domain gap, we propose to generate maps in a coarse-to-fine way. To well handle the level-dependent content differences, we design a level classifier to explore different levels of the map. Besides, we employ a map element extractor to extract the major geographic element features from satellite imageries, which is helpful to keep the main content consistency. And we design a multi-level fusion generator to generate a consistent multi-level map from the multi-level preliminary map, which further ensures the main content consistency. In addition, we collect a high-quality multi-level dataset for satellite imagery based multi-level map generation. Experimental results show that the proposed method can provide substantial improvements over the state-of-the-art alternatives in terms of both objective metric and visual quality.

While Dynamic Convolution (DY-Conv) has shown promising performance by enabling adaptive weight selection through multiple parallel weights combined with an attention mechanism, the frequency response of these weights tends to exhibit high similarity, resulting in high parameter costs but limited adaptability. In this work, we introduce Frequency Dynamic Convolution (FDConv), a novel approach that mitigates these limitations by learning a fixed parameter budget in the Fourier domain. FDConv divides this budget into frequency-based groups with disjoint Fourier indices, enabling the construction of frequency-diverse weights without increasing the parameter cost. To further enhance adaptability, we propose Kernel Spatial Modulation (KSM) and Frequency Band Modulation (FBM). KSM dynamically adjusts the frequency response of each filter at the spatial level, while FBM decomposes weights into distinct frequency bands in the frequency domain and modulates them dynamically based on local content. Extensive experiments on object detection, segmentation, and classification validate the effectiveness of FDConv. We demonstrate that when applied to ResNet-50, FDConv achieves superior performance with a modest increase of +3.6M parameters, outperforming previous methods that require substantial increases in parameter budgets (e.g., CondConv +90M, KW +76.5M). Moreover, FDConv seamlessly integrates into a variety of architectures, including ConvNeXt, Swin-Transformer, offering a flexible and efficient solution for modern vision tasks. The code is made publicly available at https://github.com/Linwei-Chen/FDConv.