Salient object detection (SOD), which simulates the human visual perception system to locate the most attractive object(s) in a scene, has been widely applied to various computer vision tasks. Now, with the advent of depth sensors, depth maps with affluent spatial information that can be beneficial in boosting the performance of SOD, can easily be captured. Although various RGB-D based SOD models with promising performance have been proposed over the past several years, an in-depth understanding of these models and challenges in this topic remains lacking. In this paper, we provide a comprehensive survey of RGB-D based SOD models from various perspectives, and review related benchmark datasets in detail. Further, considering that the light field can also provide depth maps, we review SOD models and popular benchmark datasets from this domain as well. Moreover, to investigate the SOD ability of existing models, we carry out a comprehensive evaluation, as well as attribute-based evaluation of several representative RGB-D based SOD models. Finally, we discuss several challenges and open directions of RGB-D based SOD for future research. All collected models, benchmark datasets, source code links, datasets constructed for attribute-based evaluation, and codes for evaluation will be made publicly available at https://github.com/taozh2017/RGBDSODsurvey
Weakly supervised object detection (WSOD) is a challenging task, in which image-level labels (e.g., categories of the instances in the whole image) are used to train an object detector. Many existing methods follow the standard multiple instance learning (MIL) paradigm and have achieved promising performance. However, the lack of deterministic information leads to part domination and missing instances. To address these issues, this paper focuses on identifying and fully exploiting the deterministic information in WSOD. We discover that negative instances (i.e. absolutely wrong instances), ignored in most of the previous studies, normally contain valuable deterministic information. Based on this observation, we here propose a negative deterministic information (NDI) based method for improving WSOD, namely NDI-WSOD. Specifically, our method consists of two stages: NDI collecting and exploiting. In the collecting stage, we design several processes to identify and distill the NDI from negative instances online. In the exploiting stage, we utilize the extracted NDI to construct a novel negative contrastive learning mechanism and a negative guided instance selection strategy for dealing with the issues of part domination and missing instances, respectively. Experimental results on several public benchmarks including VOC 2007, VOC 2012 and MS COCO show that our method achieves satisfactory performance.
We introduce a highly performant 3D object detector for point clouds using the DETR framework. The prior attempts all end up with suboptimal results because they fail to learn accurate inductive biases from the limited scale of training data. In particular, the queries often attend to points that are far away from the target objects, violating the locality principle in object detection. To address the limitation, we introduce a novel 3D Vertex Relative Position Encoding (3DV-RPE) method which computes position encoding for each point based on its relative position to the 3D boxes predicted by the queries in each decoder layer, thus providing clear information to guide the model to focus on points near the objects, in accordance with the principle of locality. In addition, we systematically improve the pipeline from various aspects such as data normalization based on our understanding of the task. We show exceptional results on the challenging ScanNetV2 benchmark, achieving significant improvements over the previous 3DETR in $\rm{AP}_{25}$/$\rm{AP}_{50}$ from 65.0\%/47.0\% to 77.8\%/66.0\%, respectively. In addition, our method sets a new record on ScanNetV2 and SUN RGB-D datasets.Code will be released at http://github.com/yichaoshen-MS/V-DETR.
The scattering of light observed through the turbid underwater channel is often regarded as the leading challenge when designing underwater electro-optical imaging systems. There have been many approaches to address the effects of scattering such as using pulsed laser sources to reject scattered light temporally, or using intensity modulated waveforms and matched filters to remove the scattered light spectrally. In this paper, a new method is proposed which primarily uses the backscattering asymmetry property for object detection and geometric profiling. In our approach, two parallel and identical continuous wave (CW) laser beams with narrow beam widths (~2mm) are used as active illumination sources. The two beams also have controllable spacing and aiming angle, as well as initial phase difference for convenience of scanning and profiling a target. Through theory and experimental results, it will be shown that when an object leans or tilts towards one of the beam's central trajectory, the asymmetry in the backscattered signals can be used to indicate the location or slope of the target's surface, respectively. By varying the spacing or aiming angle of the two beams, a number of surface samples can be collected to reconstruct the object's shape geometrically. The resolution and range limit of our approach are also measured and reported in this work. In application, our proposed method provides an economic solution to perform imaging through turbid underwater environments. Additionally, the idea can be combined with the pulsed or modulated laser signals for enhanced imaging results.
The spread of the novel coronavirus (COVID-19) has caused trillions of dollars in damages to the governments and health authorities by affecting the global economies. The purpose of this study is to introduce a connected smart paradigm that not only detects the possible spread of viruses but also helps to restart businesses/economies, and resume social life. We are proposing a connected Internet of Things ( IoT) based paradigm that makes use of object detection based on convolution neural networks (CNN), smart wearable and connected e-health to avoid current and future outbreaks. First, connected surveillance cameras feed continuous video stream to the server where we detect the inter-object distance to identify any social distancing violations. A violation activates area-based monitoring of active smartphone users and their current state of illness. In case a confirmed patient or a person with high symptoms is present, the system tracks exposed and infected people and appropriate measures are put into actions. We evaluated the proposed scheme for social distancing violation detection using YOLO (you only look once) v2 and v3, and for infection spread tracing using Python simulation.
Feature pyramid network (FPN) is a critical component in modern object detection frameworks. The performance gain in most of the existing FPN variants is mainly attributed to the increase of computational burden. An attempt to enhance the FPN is enriching the spatial information by expanding the receptive fields, which is promising to largely improve the detection accuracy. In this paper, we first investigate how expanding the receptive fields affect the accuracy and computational costs of FPN. We explore a baseline model called inception FPN in which each lateral connection contains convolution filters with different kernel sizes. Moreover, we point out that not all objects need such a complicated calculation and propose a new dynamic FPN (DyFPN). The output features of DyFPN will be calculated by using the adaptively selected branch according to a dynamic gating operation. Therefore, the proposed method can provide a more efficient dynamic inference for achieving a better trade-off between accuracy and computational cost. Extensive experiments conducted on MS-COCO benchmark demonstrate that the proposed DyFPN significantly improves performance with the optimal allocation of computation resources. For instance, replacing inception FPN with DyFPN reduces about 40% of its FLOPs while maintaining similar high performance.
Three-dimensional (3D) reconstruction of head Computed Tomography (CT) images elucidates the intricate spatial relationships of tissue structures, thereby assisting in accurate diagnosis. Nonetheless, securing an optimal head CT scan without deviation is challenging in clinical settings, owing to poor positioning by technicians, patient's physical constraints, or CT scanner tilt angle restrictions. Manual formatting and reconstruction not only introduce subjectivity but also strain time and labor resources. To address these issues, we propose an efficient automatic head CT images 3D reconstruction method, improving accuracy and repeatability, as well as diminishing manual intervention. Our approach employs a deep learning-based object detection algorithm, identifying and evaluating orbitomeatal line landmarks to automatically reformat the images prior to reconstruction. Given the dearth of existing evaluations of object detection algorithms in the context of head CT images, we compared ten methods from both theoretical and experimental perspectives. By exploring their precision, efficiency, and robustness, we singled out the lightweight YOLOv8 as the aptest algorithm for our task, with an mAP of 92.77% and impressive robustness against class imbalance. Our qualitative evaluation of standardized reconstruction results demonstrates the clinical practicability and validity of our method.
Image-based salient object detection (SOD) has been extensively explored in the past decades. However, SOD on 360$^\circ$ omnidirectional images is less studied owing to the lack of datasets with pixel-level annotations. Toward this end, this paper proposes a 360$^\circ$ image-based SOD dataset that contains 500 high-resolution equirectangular images. We collect the representative equirectangular images from five mainstream 360$^\circ$ video datasets and manually annotate all objects and regions over these images with precise masks with a free-viewpoint way. To the best of our knowledge, it is the first public available dataset for salient object detection on 360$^\circ$ scenes. By observing this dataset, we find that distortion from projection, large-scale complex scene and small salient objects are the most prominent characteristics. Inspired by these foundings, this paper proposes a baseline model for SOD on equirectangular images. In the proposed approach, we construct a distortion-adaptive module to deal with the distortion caused by the equirectangular projection. In addition, a multi-scale contextual integration block is introduced to perceive and distinguish the rich scenes and objects in omnidirectional scenes. The whole network is organized in a progressively manner with deep supervision. Experimental results show the proposed baseline approach outperforms the top-performanced state-of-the-art methods on 360$^\circ$ SOD dataset. Moreover, benchmarking results of the proposed baseline approach and other methods on 360$^\circ$ SOD dataset show the proposed dataset is very challenging, which also validate the usefulness of the proposed dataset and approach to boost the development of SOD on 360$^\circ$ omnidirectional scenes.
Despite the rapid advancement of object detection algorithms, processing high-resolution images on embedded devices remains a significant challenge. Theoretically, the fully convolutional network architecture used in current real-time object detectors can handle all input resolutions. However, the substantial computational demands required to process high-resolution images render them impractical for real-time applications. To address this issue, real-time object detection models typically downsample the input image for inference, leading to a loss of detail and decreased accuracy. In response, we developed Octave-YOLO, designed to process high-resolution images in real-time within the constraints of embedded systems. We achieved this through the introduction of the cross frequency partial network (CFPNet), which divides the input feature map into low-resolution, low-frequency, and high-resolution, high-frequency sections. This configuration enables complex operations such as convolution bottlenecks and self-attention to be conducted exclusively on low-resolution feature maps while simultaneously preserving the details in high-resolution maps. Notably, this approach not only dramatically reduces the computational demands of convolution tasks but also allows for the integration of attention modules, which are typically challenging to implement in real-time applications, with minimal additional cost. Additionally, we have incorporated depthwise separable convolution into the core building blocks and downsampling layers to further decrease latency. Experimental results have shown that Octave-YOLO matches the performance of YOLOv8 while significantly reducing computational demands. For example, in 1080x1080 resolution, Octave-YOLO-N is 1.56 times faster than YOLOv8, achieving nearly the same accuracy on the COCO dataset with approximately 40 percent fewer parameters and FLOPs.
The problem of identifying the 3D pose of a known object from a given 2D image has important applications in Computer Vision. Our proposed method of registering a 3D model of a known object on a given 2D photo of the object has numerous advantages over existing methods. It does not require prior training, knowledge of the camera parameters, explicit point correspondences or matching features between the image and model. Unlike techniques that estimate a partial 3D pose (as in an overhead view of traffic or machine parts on a conveyor belt), our method estimates the complete 3D pose of the object. It works on a single static image from a given view under varying and unknown lighting conditions. For this purpose we derive a novel illumination-invariant distance measure between the 2D photo and projected 3D model, which is then minimised to find the best pose parameters. Results for vehicle pose detection in real photographs are presented.
We present recurrent geometry-aware neural networks that integrate visual information across multiple views of a scene into 3D latent feature tensors, while maintaining an one-to-one mapping between 3D physical locations in the world scene and latent feature locations. Object detection, object segmentation, and 3D reconstruction is then carried out directly using the constructed 3D feature memory, as opposed to any of the input 2D images. The proposed models are equipped with differentiable egomotion-aware feature warping and (learned) depth-aware unprojection operations to achieve geometrically consistent mapping between the features in the input frame and the constructed latent model of the scene. We empirically show the proposed model generalizes much better than geometryunaware LSTM/GRU networks, especially under the presence of multiple objects and cross-object occlusions. Combined with active view selection policies, our model learns to select informative viewpoints to integrate information from by "undoing" cross-object occlusions, seamlessly combining geometry with learning from experience.
3D Multi-object tracking (MOT) is crucial to autonomous systems. Recent work often uses a tracking-by-detection pipeline, where the feature of each object is extracted independently to compute an affinity matrix. Then, the affinity matrix is passed to the Hungarian algorithm for data association. A key process of this pipeline is to learn discriminative features for different objects in order to reduce confusion during data association. To that end, we propose two innovative techniques: (1) instead of obtaining the features for each object independently, we propose a novel feature interaction mechanism by introducing Graph Neural Networks; (2) instead of obtaining the features from either 2D or 3D space as in prior work, we propose a novel joint feature extractor to learn appearance and motion features from 2D and 3D space. Through experiments on the KITTI dataset, our proposed method achieves state-of-the-art 3D MOT performance. Our project website is at http://www.xinshuoweng.com/projects/GNN3DMOT.
Previous video object segmentation approaches mainly focus on using simplex solutions between appearance and motion, limiting feature collaboration efficiency among and across these two cues. In this work, we study a novel and efficient full-duplex strategy network (FSNet) to address this issue, by considering a better mutual restraint scheme between motion and appearance in exploiting the cross-modal features from the fusion and decoding stage. Specifically, we introduce the relational cross-attention module (RCAM) to achieve bidirectional message propagation across embedding sub-spaces. To improve the model's robustness and update the inconsistent features from the spatial-temporal embeddings, we adopt the bidirectional purification module (BPM) after the RCAM. Extensive experiments on five popular benchmarks show that our FSNet is robust to various challenging scenarios (e.g., motion blur, occlusion) and achieves favourable performance against existing cutting-edges both in the video object segmentation and video salient object detection tasks. The project is publicly available at: https://dpfan.net/FSNet.
Video object segmentation (VOS) is an essential part of autonomous vehicle navigation. The real-time speed is very important for the autonomous vehicle algorithms along with the accuracy metric. In this paper, we propose a semi-supervised real-time method based on the Siamese network using a new polar representation. The input of bounding boxes is initialized rather than the object masks, which are applied to the video object detection tasks. The polar representation could reduce the parameters for encoding masks with subtle accuracy loss so that the algorithm speed can be improved significantly. An asymmetric siamese network is also developed to extract the features from different spatial scales. Moreover, the peeling convolution is proposed to reduce the antagonism among the branches of the polar head. The repeated cross-correlation and semi-FPN are designed based on this idea. The experimental results on the DAVIS-2016 dataset and other public datasets demonstrate the effectiveness of the proposed method.
Autonomous driving requires various computer vision algorithms, such as object detection and tracking.Precisely-labeled datasets (i.e., objects are fully contained in bounding boxes with only a few extra pixels) are preferred for training such algorithms, so that the algorithms can detect exact locations of the objects. However, it is very time-consuming and hence expensive to generate precise labels for image sequences at scale. In this paper, we propose DeepBbox, an algorithm that corrects loose object labels into right bounding boxes to reduce human annotation efforts. We use Cityscapes dataset to show annotation efficiency and accuracy improvement using DeepBbox. Experimental results show that, with DeepBbox,we can increase the number of object edges that are labeled automatically (within 1\% error) by 50% to reduce manual annotation time.
One of the challenges in evaluating multi-object video detection, tracking and classification systems is having publically available data sets with which to compare different systems. However, the measures of performance for tracking and classification are different. Data sets that are suitable for evaluating tracking systems may not be appropriate for classification. Tracking video data sets typically only have ground truth track IDs, while classification video data sets only have ground truth class-label IDs. The former identifies the same object over multiple frames, while the latter identifies the type of object in individual frames. This paper describes an advancement of the ground truth meta-data for the DARPA Neovision2 Tower data set to allow both the evaluation of tracking and classification. The ground truth data sets presented in this paper contain unique object IDs across 5 different classes of object (Car, Bus, Truck, Person, Cyclist) for 24 videos of 871 image frames each. In addition to the object IDs and class labels, the ground truth data also contains the original bounding box coordinates together with new bounding boxes in instances where un-annotated objects were present. The unique IDs are maintained during occlusions between multiple objects or when objects re-enter the field of view. This will provide: a solid foundation for evaluating the performance of multi-object tracking of different types of objects, a straightforward comparison of tracking system performance using the standard Multi Object Tracking (MOT) framework, and classification performance using the Neovision2 metrics. These data have been hosted publically.
Identifying different objects (man and cup) is an important problem on its own, but identifying the relationship between them (holding) is critical for many real world use cases. This paper describes an approach to reduce a visual relationship detection problem to object detection problems. The method was applied to Google AI Open Images V4 Visual Relationship Track Challenge, which was held in conjunction with 2018 European Conference on Computer Vision (ECCV 2018) and it finished as a prize winner. The challenge was to build an algorithm that detects pairs of objects in particular relations: things like "woman playing guitar," "beer on table," or "dog inside car.".
The work discusses the use of machine learning algorithms for anomaly detection in medical image analysis and how the performance of these algorithms depends on the number of annotators and the quality of labels. To address the issue of subjectivity in labeling with a single annotator, we introduce a simple and effective approach that aggregates annotations from multiple annotators with varying levels of expertise. We then aim to improve the efficiency of predictive models in abnormal detection tasks by estimating hidden labels from multiple annotations and using a re-weighted loss function to improve detection performance. Our method is evaluated on a real-world medical imaging dataset and outperforms relevant baselines that do not consider disagreements among annotators.
Up-to-date High-Definition (HD) maps are essential for self-driving cars. To achieve constantly updated HD maps, we present a deep neural network (DNN), Diff-Net, to detect changes in them. Compared to traditional methods based on object detectors, the essential design in our work is a parallel feature difference calculation structure that infers map changes by comparing features extracted from the camera and rasterized images. To generate these rasterized images, we project map elements onto images in the camera view, yielding meaningful map representations that can be consumed by a DNN accordingly. As we formulate the change detection task as an object detection problem, we leverage the anchor-based structure that predicts bounding boxes with different change status categories. To the best of our knowledge, the proposed method is the first end-to-end network that tackles the high-definition map change detection task, yielding a single stage solution. Furthermore, rather than relying on single frame input, we introduce a spatio-temporal fusion module that fuses features from history frames into the current, thus improving the overall performance. Finally, we comprehensively validate our method's effectiveness using freshly collected datasets. Results demonstrate that our Diff-Net achieves better performance than the baseline methods and is ready to be integrated into a map production pipeline maintaining an up-to-date HD map.
Conventional knowledge distillation (KD) methods for object detection mainly concentrate on homogeneous teacher-student detectors. However, the design of a lightweight detector for deployment is often significantly different from a high-capacity detector. Thus, we investigate KD among heterogeneous teacher-student pairs for a wide application. We observe that the core difficulty for heterogeneous KD (hetero-KD) is the significant semantic gap between the backbone features of heterogeneous detectors due to the different optimization manners. Conventional homogeneous KD (homo-KD) methods suffer from such a gap and are hard to directly obtain satisfactory performance for hetero-KD. In this paper, we propose the HEtero-Assists Distillation (HEAD) framework, leveraging heterogeneous detection heads as assistants to guide the optimization of the student detector to reduce this gap. In HEAD, the assistant is an additional detection head with the architecture homogeneous to the teacher head attached to the student backbone. Thus, a hetero-KD is transformed into a homo-KD, allowing efficient knowledge transfer from the teacher to the student. Moreover, we extend HEAD into a Teacher-Free HEAD (TF-HEAD) framework when a well-trained teacher detector is unavailable. Our method has achieved significant improvement compared to current detection KD methods. For example, on the MS-COCO dataset, TF-HEAD helps R18 RetinaNet achieve 33.9 mAP (+2.2), while HEAD further pushes the limit to 36.2 mAP (+4.5).