End-to-end paradigms significantly improve the accuracy of various deep-learning-based computer vision models. To this end, tasks like object detection have been upgraded by replacing non-end-to-end components, such as removing non-maximum suppression by training with a set loss based on bipartite matching. However, such an upgrade is not applicable to instance segmentation, due to its significantly higher output dimensions compared to object detection. In this paper, we propose an instance segmentation Transformer, termed ISTR, which is the first end-to-end framework of its kind. ISTR predicts low-dimensional mask embeddings, and matches them with ground truth mask embeddings for the set loss. Besides, ISTR concurrently conducts detection and segmentation with a recurrent refinement strategy, which provides a new way to achieve instance segmentation compared to the existing top-down and bottom-up frameworks. Benefiting from the proposed end-to-end mechanism, ISTR demonstrates state-of-the-art performance even with approximation-based suboptimal embeddings. Specifically, ISTR obtains a 46.8/38.6 box/mask AP using ResNet50-FPN, and a 48.1/39.9 box/mask AP using ResNet101-FPN, on the MS COCO dataset. Quantitative and qualitative results reveal the promising potential of ISTR as a solid baseline for instance-level recognition. Code has been made available at: https://github.com/hujiecpp/ISTR.
Disasters, both natural and man-made, pose significant risks to human life and infrastructure, necessitating swift and efficient search and rescue (SAR) operations. Traditional SAR methods often struggle to access hazardous areas, resulting in delayed responses and increased risk to rescuers. These methods are unable to run automated simultaneous rescues which lead the operations to face challenges such as difficulty in quickly assessing damage, locating survivors, and delivering aid. This paper address the problem by proposing a autonomous swarm of drones framework which improves the response time as well as increases accessibility zone of the rescue operation. The proposed framework DROPEX, is an autonomous UAV (Unmanned Aerial Vehicle) which employs a dual-dome architecture with surveillance drones to detect individuals in distress and payload drones to deliver aid. This framework ensures rapid deployment, accurate navigation, and efficient data transmission in disaster-stricken areas while minimizing the need for manual intervention. The drones are able to recognise the victims in need using object detection models such as YOLO and Detection Transformer (DETR) with thermal vision. By using Long Range Wide Area Network (LoRaWAN) and object detection models, the drones are able to reduce the response time and increase accessibility. The main focus is on creating a robust, scalable, and economical system to enhance the speed, efficiency, and effectiveness of disaster rescue operations.
위장 객체 검출(COD)은 복잡한 환경에서 배경과 색상 및 질감이 유사하여 은폐된 객체를 식별하고 분할하는 것을 목표로 합니다. 기존의 몇몇 COD 방법들은 보완적인 RGB-D 특징을 학습하여 검출 성능을 향상시키고자 깊이 맵(depth map)을 도입하지만, 깊이 도메인에서 은폐된 객체의 양식 특화(modality-specific) 특성을 간과합니다. 이러한 문제를 해결하기 위해, 본 논문에서는 깊이 도메인에서 시각적 은폐 프로토타입을 넘어 구별 가능한 다중 양식 특징을 발굴하는 깊이 협업 네트워크인 VCP-DCN을 제안합니다. 구체적으로, VCP-DCN은 COD 작업을 위해 다중 양식 정렬, 상호작용 및 융합을 점진적으로 수행합니다. 정렬(alignment) 단계에서는 프로토타입 대조 학습을 통해 양식 일관성 및 양식 특화 RGB/깊이 프로토타입 토큰을 학습하는 분리형 프로토타입 임베딩(SPE) 모듈을 제안합니다. 나아가 상호작용(interaction) 단계에서는 양식 일관성 RGB/깊이 프로토타입 토큰과 시각적 토큰 간의 국소 반응 맵을 통해 교차 모달 특징 표현을 강화하는 다중 양식 듀얼 어텐션(MDA) 모듈을 개발합니다. 마지막으로 융합(fusion) 단계에서는 의사결정 메커니즘을 통해 RGB/깊이 특징의 기여도를 적응적으로 측정하는 깊이 적응형 주입(DAI) 모듈을 설계하며, 이는 RGB/깊이 양식 특화 프로토타입 토큰과 양식 일관성 프로토타입 토큰 간의 유사도 거리를 계산합니다. 광범위한 실험을 통해 3개의 공신력 있는 데이터셋에서 VCP-DCN의 효과를 입증합니다.
We present a unified, efficient and effective framework for point-cloud based 3D object detection. Our two-stage approach utilizes both voxel representation and raw point cloud data to exploit respective advantages. The first stage network, with voxel representation as input, only consists of light convolutional operations, producing a small number of high-quality initial predictions. Coordinate and indexed convolutional feature of each point in initial prediction are effectively fused with the attention mechanism, preserving both accurate localization and context information. The second stage works on interior points with their fused feature for further refining the prediction. Our method is evaluated on KITTI dataset, in terms of both 3D and Bird's Eye View (BEV) detection, and achieves state-of-the-arts with a 15FPS detection rate.
Visual identification of gunmen in a crowd is a challenging problem, that requires resolving the association of a person with an object (firearm). We present a novel approach to address this problem, by defining human-object interaction (and non-interaction) bounding boxes. In a given image, human and firearms are separately detected. Each detected human is paired with each detected firearm, allowing us to create a paired bounding box that contains both object and the human. A network is trained to classify these paired-bounding-boxes into human carrying the identified firearm or not. Extensive experiments were performed to evaluate effectiveness of the algorithm, including exploiting full pose of the human, hand key-points, and their association with the firearm. The knowledge of spatially localized features is key to success of our method by using multi-size proposals with adaptive average pooling. We have also extended a previously firearm detection dataset, by adding more images and tagging in extended dataset the human-firearm pairs (including bounding boxes for firearms and gunmen). The experimental results ($AP_{hold} = 78.5$) demonstrate effectiveness of the proposed method.
In a complex road traffic scene, illegal lane intrusion of pedestrians or cyclists constitutes one of the main safety challenges in autonomous driving application. In this paper, we propose a novel object-level phase space reconstruction network (PSRNet) for motion time series classification, aiming to recognize lane intrusion actions that occur 150m ahead through a monocular camera fixed on moving vehicle. In the PSRNet, the movement of pedestrians and cyclists, specifically viewed as an observable object-level dynamic process, can be reconstructed as trajectories of state vectors in a latent phase space and further characterized by a learnable Lyapunov exponent-like classifier that indicates discrimination in terms of average exponential divergence of state trajectories. Additionally, in order to first transform video inputs into one-dimensional motion time series of each object, a lane width normalization based on visual object tracking-by-detection is presented. Extensive experiments are conducted on the THU-IntrudBehavior dataset collected from real urban roads. The results show that our PSRNet could reach the best accuracy of 98.0%, which remarkably exceeds existing action recognition approaches by more than 30%.
Anomaly detection has a wide range of applications and is especially important in industrial quality inspection. Currently, many top-performing anomaly-detection models rely on feature-embedding methods. However, these methods do not perform well on datasets with large variations in object locations. Reconstruction-based methods use reconstruction errors to detect anomalies without considering positional differences between samples. In this study, a reconstruction-based method using the noise-to-norm paradigm is proposed, which avoids the invariant reconstruction of anomalous regions. Our reconstruction network is based on M-net and incorporates multiscale fusion and residual attention modules to enable end-to-end anomaly detection and localization. Experiments demonstrate that the method is effective in reconstructing anomalous regions into normal patterns and achieving accurate anomaly detection and localization. On the MPDD and VisA datasets, our proposed method achieved more competitive results than the latest methods, and it set a new state-of-the-art standard on the MPDD dataset.
Ensuring construction site safety requires accurate and real-time detection of workers' safety helmet use, despite challenges posed by cluttered environments, densely populated work areas, and hard-to-detect small or overlapping objects caused by building obstructions. This paper proposes a novel algorithm for safety helmet wearing detection, incorporating a dynamic attention within the detection head to enhance multi-scale perception. The mechanism combines feature-level attention for scale adaptation, spatial attention for spatial localization, and channel attention for task-specific insights, improving small object detection without additional computational overhead. Furthermore, a two-way fusion strategy enables bidirectional information flow, refining feature fusion through adaptive multi-scale weighting, and enhancing recognition of occluded targets. Experimental results demonstrate a 1.7% improvement in mAP@[.5:.95] compared to the best baseline while reducing GFLOPs by 11.9% on larger sizes. The proposed method surpasses existing models, providing an efficient and practical solution for real-world construction safety monitoring.
Tracking moving objects from a video sequence requires segmentation of these objects from the background image. However, getting the actual background image automatically without object detection and using only the video is difficult. In this paper, we describe a novel algorithm that generates background from real world images without foreground detection. The algorithm assumes that the background image is shown in the majority of the video. Given this simple assumption, the method described in this paper is able to accurately generate, with high probability, the background image from a video using only a small number of binary operations.
A safe and robust autonomous driving system relies on accurate perception of the environment for application-oriented scenarios. This paper proposes deployment of the three most crucial tasks (i.e., object detection, drivable area segmentation and lane detection tasks) on embedded system for self-driving operations. To achieve this research objective, multi-tasking network is utilized with a simple encoder-decoder architecture. Comprehensive and extensive comparisons for two models based on different backbone networks are performed. All training experiments are performed on server while Nvidia Jetson Xavier NX is chosen as deployment device.
The goal of this work is to replace objects in an RGB-D scene with corresponding 3D models from a library. We approach this problem by first detecting and segmenting object instances in the scene using the approach from Gupta et al. [13]. We use a convolutional neural network (CNN) to predict the pose of the object. This CNN is trained using pixel normals in images containing rendered synthetic objects. When tested on real data, it outperforms alternative algorithms trained on real data. We then use this coarse pose estimate along with the inferred pixel support to align a small number of prototypical models to the data, and place the model that fits the best into the scene. We observe a 48% relative improvement in performance at the task of 3D detection over the current state-of-the-art [33], while being an order of magnitude faster at the same time.
Explaining predictions of deep neural networks (DNNs) is an important and nontrivial task. In this paper, we propose a practical approach to interpret decisions made by a DNN object detector that has fidelity comparable to state-of-the-art methods and sufficient computational efficiency to process large datasets. Our method relies on recent theory and approximates Shapley feature importance values. We qualitatively and quantitatively show that the proposed explanation method can be used to find image features which cause failures in DNN object detection. The developed software tool combined into the "Explain to Fix" (E2X) framework has a factor of 10 higher computational efficiency than prior methods and can be used for cluster processing using graphics processing units (GPUs). Lastly, we propose a potential extension of the E2X framework where the discovered missing features can be added into training dataset to overcome failures after model retraining.
This paper aims at constructing a light-weight object detector that inputs a depth and a color image from a stereo camera. Specifically, by extending the network architecture of YOLOv3 to 3D in the middle, it is possible to output in the depth direction. In addition, Intersection over Uninon (IoU) in 3D space is introduced to confirm the accuracy of region extraction results. In the field of deep learning, object detectors that use distance information as input are actively studied for utilizing automated driving. However, the conventional detector has a large network structure, and the real-time property is impaired. The effectiveness of the detector constructed as described above is verified using datasets. As a result of this experiment, the proposed model is able to output 3D bounding boxes and detect people whose part of the body is hidden. Further, the processing speed of the model is 44.35 fps.
Compared with expensive pixel-wise annotations, image-level labels make it possible to learn semantic segmentation in a weakly-supervised manner. Within this pipeline, the class activation map (CAM) is obtained and further processed to serve as a pseudo label to train the semantic segmentation model in a fully-supervised manner. In this paper, we argue that the lost structure information in CAM limits its application in downstream semantic segmentation, leading to deteriorated predictions. Furthermore, the inconsistent class activation scores inside the same object contradicts the common sense that each region of the same object should belong to the same semantic category. To produce sharp prediction with structure information, we introduce an auxiliary semantic boundary detection module, which penalizes the deteriorated predictions. Furthermore, we adopt smoothness loss to encourage prediction inside the object to be consistent. Experimental results on the PASCAL-VOC dataset illustrate the effectiveness of the proposed solution.
Automotive radar sensors output a lot of unwanted clutter or ghost detections, whose position and velocity do not correspond to any real object in the sensor's field of view. This poses a substantial challenge for environment perception methods like object detection or tracking. Especially problematic are clutter detections that occur in groups or at similar locations in multiple consecutive measurements. In this paper, a new algorithm for identifying such erroneous detections is presented. It is mainly based on the modeling of specific commonly occurring wave propagation paths that lead to clutter. In particular, the three effects explicitly covered are reflections at the underbody of a car or truck, signals traveling back and forth between the vehicle on which the sensor is mounted and another object, and multipath propagation via specular reflection. The latter often occurs near guardrails, concrete walls or similar reflective surfaces. Each of these effects is described both theoretically and regarding a method for identifying the corresponding clutter detections. Identification is done by analyzing detections generated from a single sensor measurement only. The final algorithm is evaluated on recordings of real extra-urban traffic. For labeling, a semi-automatic process is employed. The results are promising, both in terms of performance and regarding the very low execution time. Typically, a large part of clutter is found, while only a small ratio of detections corresponding to real objects are falsely classified by the algorithm.
Domain Adaptive Object Detection (DAOD) leverages a labeled domain to learn an object detector generalizing to a novel domain free of annotations. Recent advances align class-conditional distributions by narrowing down cross-domain prototypes (class centers). Though great success,they ignore the significant within-class variance and the domain-mismatched semantics within the training batch, leading to a sub-optimal adaptation. To overcome these challenges, we propose a novel SemantIc-complete Graph MAtching (SIGMA) framework for DAOD, which completes mismatched semantics and reformulates the adaptation with graph matching. Specifically, we design a Graph-embedded Semantic Completion module (GSC) that completes mismatched semantics through generating hallucination graph nodes in missing categories. Then, we establish cross-image graphs to model class-conditional distributions and learn a graph-guided memory bank for better semantic completion in turn. After representing the source and target data as graphs, we reformulate the adaptation as a graph matching problem, i.e., finding well-matched node pairs across graphs to reduce the domain gap, which is solved with a novel Bipartite Graph Matching adaptor (BGM). In a nutshell, we utilize graph nodes to establish semantic-aware node affinity and leverage graph edges as quadratic constraints in a structure-aware matching loss, achieving fine-grained adaptation with a node-to-node graph matching. Extensive experiments verify that SIGMA outperforms existing works significantly. Our code is available at https://github.com/CityU-AIM-Group/SIGMA.
In this work, we present a methodology to shape a fisheye-specific representation space that reflects the interaction between distortion and semantic context present in this data modality. Fisheye data has the wider field of view advantage over other types of cameras, but this comes at the expense of high radial distortion. As a result, objects further from the center exhibit deformations that make it difficult for a model to identify their semantic context. While previous work has attempted architectural and training augmentation changes to alleviate this effect, no work has attempted to guide the model towards learning a representation space that reflects this interaction between distortion and semantic context inherent to fisheye data. We introduce an approach to exploit this relationship by first extracting distortion class labels based on an object's distance from the center of the image. We then shape a backbone's representation space with a weighted contrastive loss that constrains objects of the same semantic class and distortion class to be close to each other within a lower dimensional embedding space. This backbone trained with both semantic and distortion information is then fine-tuned within an object detection setting to empirically evaluate the quality of the learnt representation. We show this method leads to performance improvements by as much as 1.1% mean average precision over standard object detection strategies and .6% improvement over other state of the art representation learning approaches.
This thesis presents a framework that integrates state-of-the-art generative AI models for real-time creation of three-dimensional (3D) objects in augmented reality (AR) environments. The primary goal is to convert diverse inputs, such as images and speech, into accurate 3D models, enhancing user interaction and immersion. Key components include advanced object detection algorithms, user-friendly interaction techniques, and robust AI models like Shap-E for 3D generation. Leveraging Vision Language Models (VLMs) and Large Language Models (LLMs), the system captures spatial details from images and processes textual information to generate comprehensive 3D objects, seamlessly integrating virtual objects into real-world environments. The framework demonstrates applications across industries such as gaming, education, retail, and interior design. It allows players to create personalized in-game assets, customers to see products in their environments before purchase, and designers to convert real-world objects into 3D models for real-time visualization. A significant contribution is democratizing 3D model creation, making advanced AI tools accessible to a broader audience, fostering creativity and innovation. The framework addresses challenges like handling multilingual inputs, diverse visual data, and complex environments, improving object detection and model generation accuracy, as well as loading 3D models in AR space in real-time. In conclusion, this thesis integrates generative AI and AR for efficient 3D model generation, enhancing accessibility and paving the way for innovative applications and improved user interactions in AR environments.
Visual grounding focuses on detecting objects from images based on language expressions. Recent Large Vision-Language Models (LVLMs) have significantly advanced visual grounding performance by training large models with large-scale datasets. However, the problem remains challenging, especially when similar objects appear in the input image. For example, an LVLM may not be able to differentiate Diet Coke and regular Coke in an image. In this case, if additional reference images of Diet Coke and regular Coke are available, it can help the visual grounding of similar objects. In this work, we introduce a new task named Multimodal Reference Visual Grounding (MRVG). In this task, a model has access to a set of reference images of objects in a database. Based on these reference images and a language expression, the model is required to detect a target object from a query image. We first introduce a new dataset to study the MRVG problem. Then we introduce a novel method, named MRVG-Net, to solve this visual grounding problem. We show that by efficiently using reference images with few-shot object detection and using Large Language Models (LLMs) for object matching, our method achieves superior visual grounding performance compared to the state-of-the-art LVLMs such as Qwen2.5-VL-7B. Our approach bridges the gap between few-shot detection and visual grounding, unlocking new capabilities for visual understanding. Project page with our code and dataset: https://irvlutd.github.io/MultiGrounding
In this work, we propose a disruptively frugal LiDAR perception dataflow that generates rather than senses parts of the environment that are either predictable based on the extensive training of the environment or have limited consequence to the overall prediction accuracy. Therefore, the proposed methodology trades off sensing energy with training data for low-power robotics and autonomous navigation to operate frugally with sensors, extending their lifetime on a single battery charge. Our proposed generative pre-training strategy for this purpose, called as radially masked autoencoding (R-MAE), can also be readily implemented in a typical LiDAR system by selectively activating and controlling the laser power for randomly generated angular regions during on-field operations. Our extensive evaluations show that pre-training with R-MAE enables focusing on the radial segments of the data, thereby capturing spatial relationships and distances between objects more effectively than conventional procedures. Therefore, the proposed methodology not only reduces sensing energy but also improves prediction accuracy. For example, our extensive evaluations on Waymo, nuScenes, and KITTI datasets show that the approach achieves over a 5% average precision improvement in detection tasks across datasets and over a 4% accuracy improvement in transferring domains from Waymo and nuScenes to KITTI. In 3D object detection, it enhances small object detection by up to 4.37% in AP at moderate difficulty levels in the KITTI dataset. Even with 90% radial masking, it surpasses baseline models by up to 5.59% in mAP/mAPH across all object classes in the Waymo dataset. Additionally, our method achieves up to 3.17% and 2.31% improvements in mAP and NDS, respectively, on the nuScenes dataset, demonstrating its effectiveness with both single and fused LiDAR-camera modalities. https://github.com/sinatayebati/Radial_MAE.