Although much significant progress has been made in the research field of object detection with deep learning, there still exists a challenging task for the objects with small size, which is notably pronounced in UAV-captured images. Addressing these issues, it is a critical need to explore the feature extraction methods that can extract more sufficient feature information of small objects. In this paper, we propose a novel method called Dense Multiscale Feature Fusion Pyramid Networks(DMFFPN), which is aimed at obtaining rich features as much as possible, improving the information propagation and reuse. Specifically, the dense connection is designed to fully utilize the representation from the different convolutional layers. Furthermore, cascade architecture is applied in the second stage to enhance the localization capability. Experiments on the drone-based datasets named VisDrone-DET suggest a competitive performance of our method.
Understanding the context of complex and cluttered scenes is a challenging problem for semantic segmentation. However, it is difficult to model the context without prior and additional supervision because the scene's factors, such as the scale, shape, and appearance of objects, vary considerably in these scenes. To solve this, we propose to learn the structures of objects and the hierarchy among objects because context is based on these intrinsic properties. In this study, we design novel hierarchical, contextual, and multiscale pyramidal representations to capture the properties from an input…
LAF-YOLOv10 with Partial Convolution Backbone, Attention-Guided Feature Pyramid, Auxiliary P2 Head, and Wise-IoU Loss for Small Object Detection in Drone Aerial Imagery
Department of Information Technology（信息科技系）；Department of Computer Science and Engineering（计算机科学与工程系）
Convolutional Neural Networks (CNNs) can provide accurate object classification. They can be extended to perform object detection by iterating over dense or selected proposed object regions. However, the runtime of such detectors scales as the total number and/or area of regions to examine per image, and training such detectors may be prohibitively slow. However, for some CNN classifier topologies, it is possible to share significant work among overlapping regions to be classified. This paper presents DenseNet, an open source system that computes dense, multiscale features from the convolution…
Nebula AI Group, School of Computer Science, Fudan University（复旦大学计算机科学技术学院星云人工智能组）；Shanghai Key Lab of Intelligent Information Processing（上海市智能信息处理重点实验室）；Department of Computer Science, The University of Hong Kong（香港大学计算机科学系）
Feature pyramids have been proven powerful in image understanding tasks that require multi-scale features. State-of-the-art methods for multi-scale feature learning focus on performing feature interactions across space and scales using neural networks with a fixed topology. In this paper, we propose graph feature pyramid networks that are capable of adapting their topological structures to varying intrinsic image structures and supporting simultaneous feature interactions across all scales. We first define an image-specific superpixel hierarchy for each input image to represent its intrinsic i…
LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images
School of Medical Instruments, Shanghai University of Medicine and Health Sciences（上海健康医学院医疗器械学院）；School of Health Sciences and Engineering, University of Shanghai for Science and Technology（上海理工大学健康科学与工程学院）；Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University（上海交通大学医学院附属上海儿童医学中心）；School of Communication & Electronic Engineering, East China Normal University（华东师范大学通信与电子工程学院）；Institute of Seismology, China Earthquake Administration（中国地震局地震研究所）
Learning Multiscale Consistency for Self-supervised Electron Microscopy Instance Segmentation
Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合性国家科学中心人工智能研究院）；University of Science and Technology of China（中国科学技术大学）
Decoupling Ego-Motion from Target Dynamics via Dual-Interval Motion Cues for UAV Detection
Department of Robotics, School of Advanced Manufacturing and Robotics（机器人学院，先进制造与机器人学院）；State Key Laboratory of Turbulence and Complex Systems（湍流与复杂系统国家重点实验室）；Peking University（北京大学）；Great Bay University（大湾大学）
West Virginia University（西弗吉尼亚大学）；State University of New York at Buffalo（纽约州立大学布法罗分校）
Object Re-IDentification (ReID), one of the most significant problems in biometrics and surveillance systems, has been extensively studied by image processing and computer vision communities in the past decades. Learning a robust and discriminative feature representation is a crucial challenge for object ReID. The problem is even more challenging in ReID based on Unmanned Aerial Vehicle (UAV) as the images are characterized by continuously varying camera parameters (e.g., view angle, altitude, etc.) of a flying drone. To address this challenge, multiscale feature representation has been consid…
Unmanned Aerial Vehicles (UAVs) especially drones, equipped with vision techniques have become very popular in recent years, with their extensive use in wide range of applications. Many of these applications require use of computer vision techniques, particularly object detection from the information captured by on-board camera. In this paper, we propose an end to end object detection model running on a UAV platform which is suitable for real-time applications. We propose a deep feature pyramid architecture which makes use of inherent properties of features extracted from Convolutional Network…
AeroLite-MDNet: Lightweight Multi-task Deviation Detection Network for UAV Landing
Institute of Cyberspace Security, Zhejiang University of Technology（浙江工业大学网络安全学院）；College of Computer Science and Technology, Zhejiang University of Technology（浙江工业大学计算机科学与技术学院）；College of Geoinformatics, Zhejiang University of Technology（浙江工业大学地理信息学院）；Binjiang Institute of Artificial Intelligence, ZJUT（滨江人工智能研究院）；Quzhou Southeast Digital Economic Development Institute（庆州东南数字经济研究院）
We propose a multigrid extension of convolutional neural networks (CNNs). Rather than manipulating representations living on a single spatial grid, our network layers operate across scale space, on a pyramid of grids. They consume multigrid inputs and produce multigrid outputs; convolutional filters themselves have both within-scale and cross-scale extent. This aspect is distinct from simple multiscale designs, which only process the input at different scales. Viewed in terms of information flow, a multigrid network passes messages across a spatial pyramid. As a consequence, receptive field si…
With the increasing demand for search and rescue, it is highly demanded to detect objects of interest in large-scale images captured by Unmanned Aerial Vehicles (UAVs), which is quite challenging due to extremely small scales of objects. Most existing methods employed Feature Pyramid Network (FPN) to enrich shallow layers' features by combing deep layers' contextual features. However, under the limitation of the inconsistency in gradient computation across different layers, the shallow layers in FPN are not fully exploited to detect tiny objects. In this paper, we propose a Scale Selection Pyr…
Recent object tracking methods depend upon deep networks or convoluted architectures. Most of those trackers can hardly meet real-time processing requirements on mobile platforms with limited computing resources. In this work, we introduce the Siamese Transformer Pyramid Network (SiamTPN), which inherits the advantages from both CNN and Transformer architectures. Specifically, we exploit the inherent feature pyramid of a lightweight network (ShuffleNetV2) and reinforce it with a Transformer to construct a robust target-specific appearance model. A centralized architecture with lateral cross at…
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Horizon Robotics（地平线机器人）
Scale variation remains a challenging problem for object detection. Common paradigms usually adopt multiscale training & testing (image pyramid) or FPN (feature pyramid network) to process objects in a wide scale range. However, multi-scale methods aggravate more variations of scale that even deep convolution neural networks with FPN cannot handle well. In this work, we propose an innovative paradigm called Instance Scale Normalization (ISN) to resolve the above problem. ISN compresses the scale space of objects into a consistent range (ISN range), in both training and testing phases. This rea…
INSA Centre Val de Loire（卢瓦尔河谷国立应用科学学院）；PRISME（PRISME实验室）
This paper addresses the problem of liver cancer segmentation in Whole Slide Image (WSI). We propose a multi-scale image processing method based on automatic end-to-end deep neural network algorithm for segmentation of cancer area. A seven-levels gaussian pyramid representation of the histopathological image was built to provide the texture information in different scales. In this work, several neural architectures were compared using the original image level for the training procedure. The proposed method is based on U-Net applied to seven levels of various resolutions (pyramidal subsumpling)…
Model efficiency has become increasingly important in computer vision. In this paper, we systematically study neural network architecture design choices for object detection and propose several key optimizations to improve efficiency. First, we propose a weighted bi-directional feature pyramid network (BiFPN), which allows easy and fast multiscale feature fusion; Second, we propose a compound scaling method that uniformly scales the resolution, depth, and width for all backbone, feature network, and box/class prediction networks at the same time. Based on these optimizations and better backbon…
Jet Propulsion Laboratory, California Institute of Technology（加州理工学院喷气推进实验室）
The next generation of Mars rotorcrafts requires on-board autonomous hazard avoidance landing. To this end, this work proposes a system that performs continuous multi-resolution height map reconstruction and safe landing spot detection. Structure-from-Motion measurements are aggregated in a pyramid structure using a novel Optimal Mixture of Gaussians formulation that provides a comprehensive uncertainty model. Our multiresolution pyramid is built more efficiently and accurately than past work by decoupling pyramid filling from the measurement updates of different resolutions. To detect the saf…
High-Frequency Semantics and Geometric Priors for End-to-End Detection Transformers in Challenging UAV Imagery
College of Mathematics and Informatics（数学与信息学院）；South China Agricultural University（华南农业大学）；Key Laboratory of Smart Agricultural Technology in Tropical South China, Ministry of Agriculture and Rural Affairs（热带南中国智能农业技术重点实验室，农业农村部）；School of Economics and Trade（经济贸易学院）
We introduce a new neural signal model designed for efficient high-resolution representation of large-scale signals. The key innovation in our multiscale implicit neural representation (MINER) is an internal representation via a Laplacian pyramid, which provides a sparse multiscale decomposition of the signal that captures orthogonal parts of the signal across scales. We leverage the advantages of the Laplacian pyramid by representing small disjoint patches of the pyramid at each scale with a small MLP. This enables the capacity of the network to adaptively increase from coarse to fine scales,…
LeafInst - Unified Instance Segmentation Network for Fine-Grained Forestry Leaf Phenotype Analysis: A New UAV based Benchmark
organization= College of Information Science；Technology \& Artificial Intelligence, Nanjing Forestry University , city= Nanjing , postcode= 210037 , state= Jiangsu , country= China；organization= Department of Geography, College of Natural Resources；organization= Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University , city= Nanjing , postcode= 210037 , state= Jiangsu , country= China；organization= School of Architecture \& Urban Planning, Shenzhen University , city= Shenzhen , postcode= 518060 , state= Guangdong , country= China；organization= School of Software, Hefei University of Technology , city= Hefei , postcode= 230601 , state= Anhui , country= China
École de Technologie Supérieure, University of Québec（魁北克大学高等技术学院）；Instituto Federal do Sul de Minas Gerais（南米纳斯吉拉斯联邦学院）；Pontificia Universidade Catolica do Paraná（巴拉那天主教大学）
Information from an image occurs over multiple and distinct spatial scales. Image pyramid multiresolution representations are a useful data structure for image analysis and manipulation over a spectrum of spatial scales. This paper employs the Gaussian-Laplacian pyramid to treat different spatial frequency bands of a texture separately. First, we generate three images corresponding to three levels of the Gaussian-Laplacian pyramid for an input image to capture intrinsic details. Then we aggregate features extracted from gray and color texture images using bio-inspired texture descriptors, info…
Discrete energy minimization is a ubiquitous task in computer vision, yet is NP-hard in most cases. In this work we propose a multiscale framework for coping with the NP-hardness of discrete optimization. Our approach utilizes algebraic multiscale principles to efficiently explore the discrete solution space, yielding improved results on challenging, non-submodular energies for which current methods provide unsatisfactory approximations. In contrast to popular multiscale methods in computer vision, that builds an image pyramid, our framework acts directly on the energy to construct an energy p…
The Chinese University of Hong Kong（香港中文大学）；Minjiang University（闽江学院）
Automatic crack detection and segmentation play a significant role in the whole system of unmanned aerial vehicle inspections. In this paper, we have implemented a deep learning framework for crack detection based on classical network architectures including Alexnet, VGG, and Resnet. Moreover, inspired by the feature pyramid network architecture, a hierarchical convolutional neural network (CNN) deep learning framework which is efficient in crack segmentation is also proposed, and its performance of it is compared with other state-of-the-art network architecture. We have summarized the existin…
Institute of Software Chinese Academy of Sciences（中国科学院软件研究所）；University of Chinese Academy of Sciences（中国科学院大学）；University at Albany, SUNY（纽约州立大学奥尔巴尼分校）；JD Digits（京东数科）
Object detection and counting are related but challenging problems, especially for drone based scenes with small objects and cluttered background. In this paper, we propose a new Guided Attention Network (GANet) to deal with both object detection and counting tasks based on the feature pyramid. Different from the previous methods relying on unsupervised attention modules, we fuse different scales of feature maps by using the proposed weakly-supervised Background Attention (BA) between the background and objects for more semantic feature representation. Then, the Foreground Attention (FA) modul…
T-SiamTPN: Temporal Siamese Transformer Pyramid Networks for Robust and Efficient UAV Tracking
Department of Electrical Engineering, Amirkabir University of Technology (AUT)（电气工程系，阿米尔卡比尔技术大学）；Qatar Computing Research Institute Hamad Bin Khalifa University（卡塔尔计算研究所，哈马德·本·卡西姆大学）
UAVD-Mamba: Deformable Token Fusion Vision Mamba for Multimodal UAV Detection
College of Mechanical and Vehicle Engineering, Hunan University（湖南大学机械与车辆工程学院）；School of Electronic Information and Communications, Huazhong University of Science and Technology（华中科技大学电子信息与通信学院）
In this technical report, we describe our solution to temporal action proposal (task 1) in ActivityNet Challenge 2019. First, we fine-tune a ResNet-50-C3D CNN on ActivityNet v1.3 based on Kinetics pretrained model to extract snippet-level video representations and then we design a Relation-Aware Pyramid Network (RapNet) to generate temporal multiscale proposals with confidence score. After that, we employ a two-stage snippet-level boundary adjustment scheme to re-rank the order of generated proposals. Ensemble methods are also been used to improve the performance of our solution, which helps u…
Cancer Hospital, Academy of Medical Sciences（中国医学科学院肿瘤医院）
This paper proposes a new convolutional neural network with multiscale processing for detecting ground-glass opacity (GGO) nodules in 3D computed tomography (CT) images, which is referred to as PiaNet for short. PiaNet consists of a feature-extraction module and a prediction module. The former module is constructed by introducing pyramid multiscale source connections into a contracting-expanding structure. The latter module includes a bounding-box regressor and a classifier that are employed to simultaneously recognize GGO nodules and estimate bounding boxes at multiple scales. To train the pr…
In this paper, we investigate the Chinese font synthesis problem and propose a Pyramid Embedded Generative Adversarial Network (PEGAN) to automatically generate Chinese character images. The PEGAN consists of one generator and one discriminator. The generator is built using one encoder-decoder structure with cascaded refinement connections and mirror skip connections. The cascaded refinement connections embed a multiscale pyramid of downsampled original input into the encoder feature maps of different layers, and multi-scale feature maps from the encoder are connected to the corresponding feat…
Feature-Augmented Deep Networks for Multiscale Building Segmentation in High-Resolution UAV and Satellite Imagery
Center for Geospatial Research, Department of Geography University of Georgia（地理空间研究中心，地理系，佐治亚大学）；Photogrammetry and Remote Sensing Department Indian Institute of Remote Sensing（摄影测量与遥感系，印度遥感研究所）；SEAI Lab, Department of Geography and the Environment University of Texas at Austin（SEAI实验室，地理与环境系，德克萨斯大学奥斯汀分校）
Facebook AI Research（Facebook AI研究院）；UC Berkeley（加州大学伯克利分校）
We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, …
Bioinformatics Institute（生物信息学研究所）；Universitat Rovira i Virgili（罗维拉-威尔吉利大学）
In this paper, we propose an efficient blood vessel segmentation method for the eye fundus images using adversarial learning with multiscale features and kernel factorization. In the generator network of the adversarial framework, spatial pyramid pooling, kernel factorization and squeeze excitation block are employed to enhance the feature representation in spatial domain on different scales with reduced computational complexity. In turn, the discriminator network of the adversarial framework is formulated by combining convolutional layers with an additional squeeze excitation block to differe…
MSDS: Deep Structural Similarity with Multiscale Representation
Fujian Key Lab for Intelligent Processing and Wireless Communication of Media Information, College of Physics and Information Engineering, Fuzhou University（福建省智能信息处理与无线通信重点实验室，物理与信息工程学院，福州大学）；School of Computing and Information Science, Fuzhou Institute of Technology（计算机与信息科学学院，福州理工学院）
ChangeDINO: DINOv3-Driven Building Change Detection in Optical Remote Sensing Imagery
1 Inst.\ of Data Science, National Cheng Kung University, \ , Taiwan - 2 Inst.\ of Intelligent Systems College of Artificial Intelligence, \ Yang Ming Chiao Tung University, Hsinchu, Taiwan
NeuroDETR: Multiscale Discrete Spatiotemporal Tokenization for Asynchronous EEG Event Detection
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Zhongguancun Academy（中关村学院）；Wuhan University（武汉大学）
Market financial forecasting is a trending area in deep learning. Deep learning models are capable of tackling the classic challenges in stock market data, such as its extremely complicated dynamics as well as long-term temporal correlation. To capture the temporal relationship among these time series, recurrent neural networks are employed. However, it is difficult for recurrent models to learn to keep track of long-term information. Convolutional Neural Networks have been utilized to better capture the dynamics and extract features for both short- and long-term forecasting. However, semantic…
TextDoctor: Unified Document Image Inpainting via Patch Pyramid Diffusion Models
College of Computer Science and Artificial Intelligence, Wenzhou University（温州大学计算机与人工智能学院）；University of British Columbia（不列颠哥伦比亚大学）；Nasdaq（纳斯达克）；Memorial University of Newfoundland（纽芬兰纪念大学）
