Arbitrary-oriented object detection (AOOD) is a challenging task to detect objects in the wild with arbitrary orientations and cluttered arrangements. Existing approaches are mainly based on anchor-based boxes or dense points, which rely on complicated hand-designed processing steps and inductive bias, such as anchor generation, transformation, and non-maximum suppression reasoning. Recently, the emerging transformer-based approaches view object detection as a direct set prediction problem that effectively removes the need for hand-designed components and inductive biases. In this paper, we propose an Arbitrary-Oriented Object DEtection TRansformer framework, termed AO2-DETR, which comprises three dedicated components. More precisely, an oriented proposal generation mechanism is proposed to explicitly generate oriented proposals, which provides better positional priors for pooling features to modulate the cross-attention in the transformer decoder. An adaptive oriented proposal refinement module is introduced to extract rotation-invariant region features and eliminate the misalignment between region features and objects. And a rotation-aware set matching loss is used to ensure the one-to-one matching process for direct set prediction without duplicate predictions. Our method considerably simplifies the overall pipeline and presents a new AOOD paradigm. Comprehensive experiments on several challenging datasets show that our method achieves superior performance on the AOOD task.
The effectiveness of Object Detection, one of the central problems in computer vision tasks, highly depends on the definition of the loss function - a measure of how accurately your ML model can predict the expected outcome. Conventional object detection loss functions depend on aggregation of metrics of bounding box regression such as the distance, overlap area and aspect ratio of the predicted and ground truth boxes (i.e. GIoU, CIoU, ICIoU etc). However, none of the methods proposed and used to date considers the direction of the mismatch between the desired ground box and the predicted, "experimental" box. This shortage results in slower and less effective convergence as the predicted box can "wander around" during the training process and eventually end up producing a worse model. In this paper a new loss function SIoU was suggested, where penalty metrics were redefined considering the angle of the vector between the desired regression. Applied to conventional Neural Networks and datasets it is shown that SIoU improves both the speed of training and the accuracy of the inference. The effectiveness of the proposed loss function was revealed in a number of simulations and tests.
Privacy protection is a crucial issue in machine learning algorithms, and the current privacy protection is combined with traditional artificial neural networks based on real values. Spiking neural network (SNN), the new generation of artificial neural networks, plays a crucial role in many fields. Therefore, research on the privacy protection of SNN is urgently needed. This paper combines the differential privacy(DP) algorithm with SNN and proposes a differentially private spiking neural network (DPSNN). The SNN uses discrete spike sequences to transmit information, combined with the gradient noise introduced by DP so that SNN maintains strong privacy protection. At the same time, to make SNN maintain high performance while obtaining high privacy protection, we propose the temporal enhanced pooling (TEP) method. It fully integrates the temporal information of SNN into the spatial information transfer, which enables SNN to perform better information transfer. We conduct experiments on static and neuromorphic datasets, and the experimental results show that our algorithm still maintains high performance while providing strong privacy protection.
Recently, Transformer networks have achieved impressive results on a variety of vision tasks. However, most of them are computationally expensive and not suitable for real-world mobile applications. In this work, we present Mobile Convolutional Vision Transformer (MoCoViT), which improves in performance and efficiency by introducing transformer into mobile convolutional networks to leverage the benefits of both architectures. Different from recent works on vision transformer, the mobile transformer block in MoCoViT is carefully designed for mobile devices and is very lightweight, accomplished through two primary modifications: the Mobile Self-Attention (MoSA) module and the Mobile Feed Forward Network (MoFFN). MoSA simplifies the calculation of the attention map through Branch Sharing scheme while MoFFN serves as a mobile version of MLP in the transformer, further reducing the computation by a large margin. Comprehensive experiments verify that our proposed MoCoViT family outperform state-of-the-art portable CNNs and transformer neural architectures on various vision tasks. On ImageNet classification, it achieves 74.5% top-1 accuracy at 147M FLOPs, gaining 1.2% over MobileNetV3 with less computations. And on the COCO object detection task, MoCoViT outperforms GhostNet by 2.1 AP in RetinaNet framework.
Substantial progress has been made in the field of object detection in road scenes. However, it is mainly focused on vehicles and pedestrians. To this end, we investigate traffic cone detection, an object category crucial for road effects and maintenance. In this work, the YOLOv5 algorithm is employed, in order to find a solution for the efficient and fast detection of traffic cones. The YOLOv5 can achieve a high detection accuracy with the score of IoU up to 91.31%. The proposed method is been applied to an RGB roadwork image dataset, collected from various sources.
Monocular 3D object detection (Mono3D) has achieved tremendous improvements with emerging large-scale autonomous driving datasets and the rapid development of deep learning techniques. However, caused by severe domain gaps (e.g., the field of view (FOV), pixel size, and object size among datasets), Mono3D detectors have difficulty in generalization, leading to drastic performance degradation on unseen domains. To solve these issues, we combine the position-invariant transform and multi-scale training with the pixel-size depth strategy to construct an effective unified camera-generalized paradigm (CGP). It fully considers discrepancies in the FOV and pixel size of images captured by different cameras. Moreover, we further investigate the obstacle in quantitative metrics when cross-dataset inference through an exhaustive systematic study. We discern that the size bias of prediction leads to a colossal failure. Hence, we propose the 2D-3D geometry-consistent object scaling strategy (GCOS) to bridge the gap via an instance-level augment. Our method called DGMono3D achieves remarkable performance on all evaluated datasets and surpasses the SoTA unsupervised domain adaptation scheme even without utilizing data on the target domain.
Thermal bridges are weak areas of building envelopes that conduct more heat to the outside than surrounding envelope areas. They lead to increased energy consumption and the formation of mold. With a neural network approach, we demonstrate a method of automatically detecting thermal bridges on building rooftops from panorama drone images of whole city districts. To train the neural network, we created a dataset including 917 images and 6895 annotations. The images in the dataset contain thermal information for detecting thermal bridges and a height map for rooftop recognition in addition to regular RGB information. Due to the small dataset, our approach currently only has an average recall of 9.4% @IoU:0.5-0.95 (14.4% for large objects). Nevertheless, our approach reliably detects structures only on rooftops and not on other parts of buildings, without any additional segmentation effort of building parts.
Adverse weather conditions can negatively affect LiDAR-based object detectors. In this work, we focus on the phenomenon of vehicle gas exhaust condensation in cold weather conditions. This everyday effect can influence the estimation of object sizes, orientations and introduce ghost object detections, compromising the reliability of the state of the art object detectors. We propose to solve this problem by using data augmentation and a novel training loss term. To effectively train deep neural networks, a large set of labeled data is needed. In case of adverse weather conditions, this process can be extremely laborious and expensive. We address this issue in two steps: First, we present a gas exhaust data generation method based on 3D surface reconstruction and sampling which allows us to generate large sets of gas exhaust clouds from a small pool of labeled data. Second, we introduce a point cloud augmentation process that can be used to add gas exhaust to datasets recorded in good weather conditions. Finally, we formulate a new training loss term that leverages the augmented point cloud to increase object detection robustness by penalizing predictions that include noise. In contrast to other works, our method can be used with both grid-based and point-based detectors. Moreover, since our approach does not require any network architecture changes, inference times remain unchanged. Experimental results on real data show that our proposed method greatly increases robustness to gas exhaust and noisy data.
Accurate detection of objects in 3D point clouds is a key problem in autonomous driving systems. Collaborative perception can incorporate information from spatially diverse sensors and provide significant benefits for improving the perception accuracy of autonomous driving systems. In this work, we consider that the autonomous vehicle uses local point cloud data and combines information from neighboring infrastructures through wireless links for cooperative 3D object detection. However, information sharing among vehicle and infrastructures in predefined communication schemes may result in communication congestion and/or bring limited performance improvement. To this end, we propose a novel collaborative 3D object detection framework that consists of three components: feature learning networks that map point clouds into feature maps; an efficient communication block that propagates compact and fine-grained query feature maps from vehicle to support infrastructures and optimizes attention weights between query and key to refine support feature maps; a region proposal network that fuses local feature maps and weighted support feature maps for 3D object detection. We evaluate the performance of the proposed framework using a synthetic cooperative dataset created in two complex driving scenarios: a roundabout and a T-junction. Experiment results and bandwidth usage analysis demonstrate that our approach can save communication and computation costs and significantly improve detection performance under different detection difficulties in all scenarios.
Typical representations for arbitrary-oriented object detection tasks include oriented bounding box (OBB), quadrilateral bounding box (QBB), and point set (PointSet). Each representation encounters problems that correspond to its characteristics, such as the boundary discontinuity, square-like problem, representation ambiguity, and isolated points, which lead to inaccurate detection. Although many effective strategies have been proposed for various representations, there is still no unified solution. Current detection methods based on Gaussian modeling have demonstrated the possibility of breaking this dilemma; however, they remain limited to OBB. To go further, in this paper, we propose a unified Gaussian representation called G-Rep to construct Gaussian distributions for OBB, QBB, and PointSet, which achieves a unified solution to various representations and problems. Specifically, PointSet or QBB-based object representations are converted into Gaussian distributions, and their parameters are optimized using the maximum likelihood estimation algorithm. Then, three optional Gaussian metrics are explored to optimize the regression loss of the detector because of their excellent parameter optimization mechanisms. Furthermore, we also use Gaussian metrics for sampling to align label assignment and regression loss. Experimental results on several public available datasets, such as DOTA, HRSC2016, UCAS-AOD, and ICDAR2015, show the excellent performance of the proposed method for arbitrary-oriented object detection.
Multi-object tracking (MOT) aims to associate target objects across video frames in order to obtain entire moving trajectories. With the advancement of deep neural networks and the increasing demand for intelligent video analysis, MOT has gained significantly increased interest in the computer vision community. Embedding methods play an essential role in object location estimation and temporal identity association in MOT. Unlike other computer vision tasks, such as image classification, object detection, re-identification, and segmentation, embedding methods in MOT have large variations, and they have never been systematically analyzed and summarized. In this survey, we first conduct a comprehensive overview with in-depth analysis for embedding methods in MOT from seven different perspectives, including patch-level embedding, single-frame embedding, cross-frame joint embedding, correlation embedding, sequential embedding, tracklet embedding, and cross-track relational embedding. We further summarize the existing widely used MOT datasets and analyze the advantages of existing state-of-the-art methods according to their embedding strategies. Finally, some critical yet under-investigated areas and future research directions are discussed.
Camouflaged object detection intends to discover the concealed objects hidden in the surroundings. Existing methods follow the bio-inspired framework, which first locates the object and second refines the boundary. We argue that the discovery of camouflaged objects depends on the recurrent search for the object and the boundary. The recurrent processing makes the human tired and helpless, but it is just the advantage of the transformer with global search ability. Therefore, a dual-task interactive transformer is proposed to detect both accurate position of the camouflaged object and its detailed boundary. The boundary feature is considered as Query to improve the camouflaged object detection, and meanwhile the object feature is considered as Query to improve the boundary detection. The camouflaged object detection and the boundary detection are fully interacted by multi-head self-attention. Besides, to obtain the initial object feature and boundary feature, transformer-based backbones are adopted to extract the foreground and background. The foreground is just object, while foreground minus background is considered as boundary. Here, the boundary feature can be obtained from blurry boundary region of the foreground and background. Supervised by the object, the background and the boundary ground truth, the proposed model achieves state-of-the-art performance in public datasets. https://github.com/liuzywen/COD
Convolutional Neural Networks (ConvNets or CNNs) have been candidly deployed in the scope of computer vision and related fields. Nevertheless, the dynamics of training of these neural networks lie still elusive: it is hard and computationally expensive to train them. A myriad of architectures and training strategies have been proposed to overcome this challenge and address several problems in image processing such as speech, image and action recognition as well as object detection. In this article, we propose a novel Particle Swarm Optimization (PSO) based training for ConvNets. In such framework, the vector of weights of each ConvNet is typically cast as the position of a particle in phase space whereby PSO collaborative dynamics intertwines with Stochastic Gradient Descent (SGD) in order to boost training performance and generalization. Our approach goes as follows: i) [regular phase] each ConvNet is trained independently via SGD; ii) [collaborative phase] ConvNets share among themselves their current vector of weights (or particle-position) along with their gradient estimates of the Loss function. Distinct step sizes are coined by distinct ConvNets. By properly blending ConvNets with large (possibly random) step-sizes along with more conservative ones, we propose an algorithm with competitive performance with respect to other PSO-based approaches on Cifar-10 and Cifar-100 (accuracy of 98.31% and 87.48%). These accuracy levels are obtained by resorting to only four ConvNets -- such results are expected to scale with the number of collaborative ConvNets accordingly. We make our source codes available for download https://github.com/leonlha/PSO-ConvNet-Dynamics.
Preys in the wild evolve to be camouflaged to avoid being recognized by predators. In this way, camouflage acts as a key defence mechanism across species that is critical to survival. To detect and segment the whole scope of a camouflaged object, camouflaged object detection (COD) is introduced as a binary segmentation task, with the binary ground truth camouflage map indicating the exact regions of the camouflaged objects. In this paper, we revisit this task and argue that the binary segmentation setting fails to fully understand the concept of camouflage. We find that explicitly modeling the conspicuousness of camouflaged objects against their particular backgrounds can not only lead to a better understanding about camouflage, but also provide guidance to designing more sophisticated camouflage techniques. Furthermore, we observe that it is some specific parts of camouflaged objects that make them detectable by predators. With the above understanding about camouflaged objects, we present the first triple-task learning framework to simultaneously localize, segment, and rank camouflaged objects, indicating the conspicuousness level of camouflage. As no corresponding datasets exist for either the localization model or the ranking model, we generate localization maps with an eye tracker, which are then processed according to the instance level labels to generate our ranking-based training and testing dataset. We also contribute the largest COD testing set to comprehensively analyse performance of the COD models. Experimental results show that our triple-task learning framework achieves new state-of-the-art, leading to a more explainable COD network. Our code, data, and results are available at: \url{https://github.com/JingZhang617/COD-Rank-Localize-and-Segment}.
The global and local contexts significantly contribute to the integrity of predictions in Salient Object Detection (SOD). Unfortunately, existing methods still struggle to generate complete predictions with fine details. There are two major problems in conventional approaches: first, for global context, high-level CNN-based encoder features cannot effectively catch long-range dependencies, resulting in incomplete predictions. Second, downsampling the ground truth to fit the size of predictions will introduce inaccuracy as the ground truth details are lost during interpolation or pooling. Thus, in this work, we developed a Transformer-based network and framed a supervised task for a branch to learn the global context information explicitly. Besides, we adopt Pixel Shuffle from Super-Resolution (SR) to reshape the predictions back to the size of ground truth instead of the reverse. Thus details in the ground truth are untouched. In addition, we developed a two-stage Context Refinement Module (CRM) to fuse global context and automatically locate and refine the local details in the predictions. The proposed network can guide and correct itself based on the global and local context generated, thus is named, Self-Refined Transformer (SelfReformer). Extensive experiments and evaluation results on five benchmark datasets demonstrate the outstanding performance of the network, and we achieved the state-of-the-art.
The remarkable breakthroughs in point cloud representation learning have boosted their usage in real-world applications such as self-driving cars and virtual reality. However, these applications usually have an urgent requirement for not only accurate but also efficient 3D object detection. Recently, knowledge distillation has been proposed as an effective model compression technique, which transfers the knowledge from an over-parameterized teacher to a lightweight student and achieves consistent effectiveness in 2D vision. However, due to point clouds' sparsity and irregularity, directly applying previous image-based knowledge distillation methods to point cloud detectors usually leads to unsatisfactory performance. To fill the gap, this paper proposes PointDistiller, a structured knowledge distillation framework for point clouds-based 3D detection. Concretely, PointDistiller includes local distillation which extracts and distills the local geometric structure of point clouds with dynamic graph convolution and reweighted learning strategy, which highlights student learning on the crucial points or voxels to improve knowledge distillation efficiency. Extensive experiments on both voxels-based and raw points-based detectors have demonstrated the effectiveness of our method over seven previous knowledge distillation methods. For instance, our 4X compressed PointPillars student achieves 2.8 and 3.4 mAP improvements on BEV and 3D object detection, outperforming its teacher by 0.9 and 1.8 mAP, respectively. Codes have been released at https://github.com/RunpeiDong/PointDistiller.
Molecular structures are always depicted as 2D printed form in scientific documents like journal papers and patents. However, these 2D depictions are not machine-readable. Due to a backlog of decades and an increasing amount of these printed literature, there is a high demand for the translation of printed depictions into machine-readable formats, which is known as Optical Chemical Structure Recognition (OCSR). Most OCSR systems developed over the last three decades follow a rule-based approach where the key step of vectorization of the depiction is based on the interpretation of vectors and nodes as bonds and atoms. Here, we present a practical software MolMiner, which is primarily built up using deep neural networks originally developed for semantic segmentation and object detection to recognize atom and bond elements from documents. These recognized elements can be easily connected as a molecular graph with distance-based construction algorithm. We carefully evaluate our software on four benchmark datasets with the state-of-the-art performance. Various real application scenarios are also tested, yielding satisfactory outcomes. The free download links of Mac and Windows versions are available: Mac: https://molminer-cdn.iipharma.cn/pharma-mind/artifact/latest/mac/PharmaMind-mac-latest-setup.dmg and Windows: https://molminer-cdn.iipharma.cn/pharma-mind/artifact/latest/win/PharmaMind-win-latest-setup.exe
Unlike existing knowledge distillation methods focus on the baseline settings, where the teacher models and training strategies are not that strong and competing as state-of-the-art approaches, this paper presents a method dubbed DIST to distill better from a stronger teacher. We empirically find that the discrepancy of predictions between the student and a stronger teacher may tend to be fairly severer. As a result, the exact match of predictions in KL divergence would disturb the training and make existing methods perform poorly. In this paper, we show that simply preserving the relations between the predictions of teacher and student would suffice, and propose a correlation-based loss to capture the intrinsic inter-class relations from the teacher explicitly. Besides, considering that different instances have different semantic similarities to each class, we also extend this relational match to the intra-class level. Our method is simple yet practical, and extensive experiments demonstrate that it adapts well to various architectures, model sizes and training strategies, and can achieve state-of-the-art performance consistently on image classification, object detection, and semantic segmentation tasks. Code is available at: https://github.com/hunto/DIST_KD .
Bolt joints are very common and important in engineering structures. Due to extreme service environment and load factors, bolts often get loose or even disengaged. To real-time or timely detect the loosed or disengaged bolts is an urgent need in practical engineering, which is critical to keep structural safety and service life. In recent years, many bolt loosening detection methods using deep learning and machine learning techniques have been proposed and are attracting more and more attention. However, most of these studies use bolt images captured in laboratory for deep leaning model training. The images are obtained in a well-controlled light, distance, and view angle conditions. Also, the bolted structures are well designed experimental structures with brand new bolts and the bolts are exposed without any shelter nearby. It is noted that in practical engineering, the above well controlled lab conditions are not easy realized and the real bolt images often have blur edges, oblique perspective, partial occlusion and indistinguishable colors etc., which make the trained models obtained in laboratory conditions loss their accuracy or fails. Therefore, the aim of this study is to develop a dataset named NPU-BOLT for bolt object detection in natural scene images and open it to researchers for public use and further development. In the first version of the dataset, it contains 337 samples of bolt joints images mainly in the natural environment, with image data sizes ranging from 400*400 to 6000*4000, totaling approximately 1275 bolt targets. The bolt targets are annotated into four categories named blur bolt, bolt head, bolt nut and bolt side. The dataset is tested with advanced object detection models including yolov5, Faster-RCNN and CenterNet. The effectiveness of the dataset is validated.
Masked AutoEncoder (MAE) has recently led the trends of visual self-supervision area by an elegant asymmetric encoder-decoder design, which significantly optimizes both the pre-training efficiency and fine-tuning accuracy. Notably, the success of the asymmetric structure relies on the "global" property of Vanilla Vision Transformer (ViT), whose self-attention mechanism reasons over arbitrary subset of discrete image patches. However, it is still unclear how the advanced Pyramid-based ViTs (e.g., PVT, Swin) can be adopted in MAE pre-training as they commonly introduce operators within "local" windows, making it difficult to handle the random sequence of partial vision tokens. In this paper, we propose Uniform Masking (UM), successfully enabling MAE pre-training for Pyramid-based ViTs with locality (termed "UM-MAE" for short). Specifically, UM includes a Uniform Sampling (US) that strictly samples $1$ random patch from each $2 \times 2$ grid, and a Secondary Masking (SM) which randomly masks a portion of (usually $25\%$) the already sampled regions as learnable tokens. US preserves equivalent elements across multiple non-overlapped local windows, resulting in the smooth support for popular Pyramid-based ViTs; whilst SM is designed for better transferable visual representations since US reduces the difficulty of pixel recovery pre-task that hinders the semantic learning. We demonstrate that UM-MAE significantly improves the pre-training efficiency (e.g., it speeds up and reduces the GPU memory by $\sim 2\times$) of Pyramid-based ViTs, but maintains the competitive fine-tuning performance across downstream tasks. For example using HTC++ detector, the pre-trained Swin-Large backbone self-supervised under UM-MAE only in ImageNet-1K can even outperform the one supervised in ImageNet-22K. The codes are available at https://github.com/implus/UM-MAE.