DAC-SDC Low Power Object Detection Challenge for UAV Applications (original) (raw)

VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results

Lecture Notes in Computer Science

Recently, automatic visual data understanding from drone platforms becomes highly demanding. To facilitate the study, the Vision Meets Drone Object Detection in Image Challenge is held the second time in conjunction with the 17-th International Conference on Computer Vision (ICCV 2019), focuses on image object detection on drones. Results of 33 object detection algorithms are presented. For each participating detector, a short description is provided in the appendix. Our goal is to advance the state-of-the-art detection algorithms and provide a comprehensive evaluation platform for them. The evaluation protocol of the VisDrone-DET2019 Challenge and the comparison results of all the submitted detectors on the released dataset are publicly available at the website: http: //www.aiskyeye.com/. The results demonstrate that there still remains a large room for improvement for object detection algorithms on drones.

VisDrone-DET2019: The Vision Meets Drone Object Detection in Image Challenge Results

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), 2019

Efficient Object Detection Model for Real-Time UAV Applications

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 Networks by capturing more generic features in the images (such as edge, color etc.) along with the minute detailed features specific to the classes contained in our problem. We use VisDrone'18 dataset for our studies which contain different objects such as pedestrians, vehicles, bicycles etc. We provide software and hardware architecture of our platform used in this study. We implemented our model with both ResNet and MobileNet as convolutional bases. Our model combined with modified focal loss function, produced a desirable performance of 30.6 mAP for object detection with an inference time of 14 fps. We compared our results with RetinaNet-ResNet-50 and HAL-RetinaNet and shown that our model combined with MobileNet as backend feature extractor gave the best results in terms of accuracy, speed and memory efficiency and is best suitable for real time object detection with drones.

Efficient Object Detection Model for Real-time UAV Application

Computer and Information Science, 2021

Unmanned Aerial Vehicles (UAVs) equipped with vision capabilities have become popular in recent years. Many applications have especially been employed object detection techniques extracted from the information captured by an onboard camera. However, object detection on UAVs requires high performance, which has a negative effect on the result. In this article, we propose a deep feature pyramid architecture with a modified focal loss function, which enables it to reduce the class imbalance. Moreover, the proposed method employed an end to end object detection model running on the UAV platform for real-time application. To evaluate the proposed architecture, we combined our model with Resnet and MobileNet as a backend network, and we compared it with RetinaNet and HAL-RetinaNet. Our model produced a performance of 30.6 mAP with an inference time of 14 fps. This result shows that our proposed model outperformed RetinaNet by 6.2 mAP.

On-Board Small-Scale Object Detection for Unmanned Aerial Vehicles (UAVs)

Drones

Object detection is a critical task that becomes difficult when dealing with onboard detection using aerial images and computer vision technique. The main challenges with aerial images are small target sizes, low resolution, occlusion, attitude, and scale variations, which affect the performance of many object detectors. The accuracy of the detection and the efficiency of the inference are always trade-offs. We modified the architecture of CenterNet and used different CNN-based backbones of ResNet18, ResNet34, ResNet50, ResNet101, ResNet152, Res2Net50, Res2Net101, DLA-34, and hourglass14. A comparison of the modified CenterNet with nine CNN-based backbones is conducted and validated using three challenging datasets, i.e., VisDrone, Stanford Drone dataset (SSD), and AU-AIR. We also implemented well-known off-the-shelf object detectors, i.e., YoloV1 to YoloV7, SSD-MobileNet-V2, and Faster RCNN. The proposed approach and state-of-the-art object detectors are optimized and then implemen...

Unmanned aerial vehicles and machine learning for detecting objects in real time

Bulletin of Electrical Engineering and Informatics

An unmanned aerial vehicle (UAV) image recognition system in real-time is proposed in this study. To begin, the you only look once (YOLO) detector has been retrained to better recognize objects in UAV photographs. The trained YOLO detector makes a trade-off between speed and precision in object recognition and localization to account for four typical moving entities caught by UAVs (cars, buses, trucks, and people). An additional 1500 UAV photographs captured by the embedded UAV camera are fed into the YOLO, which uses those probabilities to estimate the bounding box for the entire image. When it comes to object detection, the YOLO competes with other deep-learning frameworks such as the faster region convolutional neural network. The proposed system is tested on a wild test set of 1500 UAV photographs with graphics processing unit GPU acceleration, proving that it can distinguish objects in UAV images effectively and consistently in real-time at a detection speed of 60 frames per se...

VisDrone-VID2019: The Vision Meets Drone Object Detection in Video Challenge Results

2019

Video object detection has drawn great attention recently. The Vision Meets Drone Object Detection in Video Challenge 2019 (VisDrone-VID2019) is held to advance the state-of-the-art in video object detection for videos captured by drones. Specifically, there are 13 teams participating the challenge. We also report the results of 6 state-of-the-art detectors on the collected dataset. A short description is provided in the appendix for each participating detector. We present the analysis and discussion of the challenge results. Both the dataset and the challenge results are publicly available at the challenge website: http://www.aiskyeye.com/.

Object Detection for Unmanned Aerial Vehicle Camera via Convolutional Neural Networks

IEEE journal on miniaturization for air and space systems, 2021

The object tracking alongside the image segmentation have recently become of particular significance in satellite and aerial imagery. The latest achievements in this field are closely related to the application of the deep-learning algorithms and, particularly, convolutional neural networks (CNNs). Supplemented by the sufficient amount of the training data CNNs provide the advantageous performance in comparison to the classical methods based on Viola-Jones or Support vector machines. However, the application of CNNs for the object detection on the aerial images faces several general issues that cause classification error. The first one is related to the limited camera shooting angle and spatial resolution. The second one arises from the restricted dataset for specific classes of objects that rarely appear in the captured data. This paper represents a comparative study on the effectiveness of different deep neural networks for detection of the objects with similar patterns on the images within a limited amount of the pre-trained datasets. It has been revealed that YOLO ver. 3 network enables better accuracy and faster analysis than R-CNN, Fast R-CNN, Faster R-CNN, and SSD architectures. This has been demonstrated on example of "Stanford Dataset", "DOTA v-1.5", and "xView 2018 Detection" datasets. The following metrics on the accuracy have been obtained for the YOLO ver. 3 network: 89.12 mAP (Stanford Dataset), 80.20 mAP (DOTA v-1.5), and 78.29 (xView 2018) for testing; and 85.51 mAP (Stanford Dataset), 79.28 (DOTA v-1.5), and 79.92 (xView 2018) on validation with the analysis speed of 26.82 frames per second.

Embedded Real-Time Object Detection for a UAV Warning System

2017 IEEE International Conference on Computer Vision Workshops (ICCVW), 2017

In this paper, we demonstrate and evaluate a method to perform real-time object detection on-board a UAV using the state of the art YOLOv2 object detection algorithm running on an NVIDIA Jetson TX2, an GPU platform targeted at power constrained mobile applications that use neural networks under the hood. This, as a result of comparing several cutting edge object detection algorithms. Multiple evaluations we present provide insights that help choose the optimal object detection configuration given certain frame rate and detection accuracy requirements. We propose how this setup running on-board a UAV can be used to process a video feed during emergencies in real-time, and feed a decision support warning system using the generated detections.

Real-time object detection towards high power efficiency

2018 Design, Automation & Test in Europe Conference & Exhibition (DATE)

In recent years, Convolutional Neural Network (CNN) has been widely applied in computer vision tasks and has achieved significant improvement in image object detection. The CNN methods consume more computation as well as storage, so GPU is introduced for real-time object detection. However, due to the high power consumption of GPU, it is difficult to adopt GPU in mobile applications like automatic driving. The previous work proposes some optimizing techniques to lower the power consumption of object detection on mobile GPU or FPGA. In the first Low-Power Image Recognition Challenge (LPIRC), our system achieved the best result with mAP/Energy on mobile GPU platforms. We further research the acceleration of detection algorithms and implement two more systems for real-time detection on FPGA with higher energy efficiency. In this paper, we will introduce the object detection algorithms and summarize the optimizing techniques in three of our previous energy efficient detection systems on different hardware platforms for object detection.

DAC-SDC Low Power Object Detection Challenge for UAV Applications (original) (raw)

Related papers