Balanced single-shot object detection using cross-context attention-guided network

2022 ◽  
Vol 122 ◽  
pp. 108258
Author(s):  
Shuyu Miao ◽  
Shanshan Du ◽  
Rui Feng ◽  
Yuejie Zhang ◽  
Huayu Li ◽  
...  
Keyword(s):  
Symmetry ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 1718
Author(s):  
Chien-Hsing Chou ◽  
Yu-Sheng Su ◽  
Che-Ju Hsu ◽  
Kong-Chang Lee ◽  
Ping-Hsuan Han

In this study, we designed a four-dimensional (4D) audiovisual entertainment system called Sense. This system comprises a scene recognition system and hardware modules that provide haptic sensations for users when they watch movies and animations at home. In the scene recognition system, we used Google Cloud Vision to detect common scene elements in a video, such as fire, explosions, wind, and rain, and further determine whether the scene depicts hot weather, rain, or snow. Additionally, for animated videos, we applied deep learning with a single shot multibox detector to detect whether the animated video contained scenes of fire-related objects. The hardware module was designed to provide six types of haptic sensations set as line-symmetry to provide a better user experience. After the system considers the results of object detection via the scene recognition system, the system generates corresponding haptic sensations. The system integrates deep learning, auditory signals, and haptic sensations to provide an enhanced viewing experience.


2021 ◽  
Vol 7 (4) ◽  
pp. 64
Author(s):  
Tanguy Ophoff ◽  
Cédric Gullentops ◽  
Kristof Van Beeck ◽  
Toon Goedemé

Object detection models are usually trained and evaluated on highly complicated, challenging academic datasets, which results in deep networks requiring lots of computations. However, a lot of operational use-cases consist of more constrained situations: they have a limited number of classes to be detected, less intra-class variance, less lighting and background variance, constrained or even fixed camera viewpoints, etc. In these cases, we hypothesize that smaller networks could be used without deteriorating the accuracy. However, there are multiple reasons why this does not happen in practice. Firstly, overparameterized networks tend to learn better, and secondly, transfer learning is usually used to reduce the necessary amount of training data. In this paper, we investigate how much we can reduce the computational complexity of a standard object detection network in such constrained object detection problems. As a case study, we focus on a well-known single-shot object detector, YoloV2, and combine three different techniques to reduce the computational complexity of the model without reducing its accuracy on our target dataset. To investigate the influence of the problem complexity, we compare two datasets: a prototypical academic (Pascal VOC) and a real-life operational (LWIR person detection) dataset. The three optimization steps we exploited are: swapping all the convolutions for depth-wise separable convolutions, perform pruning and use weight quantization. The results of our case study indeed substantiate our hypothesis that the more constrained a problem is, the more the network can be optimized. On the constrained operational dataset, combining these optimization techniques allowed us to reduce the computational complexity with a factor of 349, as compared to only a factor 9.8 on the academic dataset. When running a benchmark on an Nvidia Jetson AGX Xavier, our fastest model runs more than 15 times faster than the original YoloV2 model, whilst increasing the accuracy by 5% Average Precision (AP).


2021 ◽  
Author(s):  
Yu Wang ◽  
Ye Zhang ◽  
Shaohua Zhai ◽  
Hao Chen ◽  
Shaoqi Shi ◽  
...  

2021 ◽  
Author(s):  
Danpei Zhao ◽  
Zhichao Yuan ◽  
Zhenwei Shi ◽  
Fengying Xie

Author(s):  
Limu Chen ◽  
Ye Xia ◽  
Dexiong Pan ◽  
Chengbin Wang

<p>Deep-learning based navigational object detection is discussed with respect to active monitoring system for anti-collision between vessel and bridge. Motion based object detection method widely used in existing anti-collision monitoring systems is incompetent in dealing with complicated and changeable waterway for its limitations in accuracy, robustness and efficiency. The video surveillance system proposed contains six modules, including image acquisition, detection, tracking, prediction, risk evaluation and decision-making, and the detection module is discussed in detail. A vessel-exclusive dataset with tons of image samples is established for neural network training and a SSD (Single Shot MultiBox Detector) based object detection model with both universality and pertinence is generated attributing to tactics of sample filtering, data augmentation and large-scale optimization, which make it capable of stable and intelligent vessel detection. Comparison results with conventional methods indicate that the proposed deep-learning method shows remarkable advantages in robustness, accuracy, efficiency and intelligence. In-situ test is carried out at Songpu Bridge in Shanghai, and the results illustrate that the method is qualified for long-term monitoring and providing information support for further analysis and decision making.</p>


2020 ◽  
Vol 12 (13) ◽  
pp. 2136 ◽  
Author(s):  
Arun Narenthiran Veeranampalayam Sivakumar ◽  
Jiating Li ◽  
Stephen Scott ◽  
Eric Psota ◽  
Amit J. Jhala ◽  
...  

Mid- to late-season weeds that escape from the routine early-season weed management threaten agricultural production by creating a large number of seeds for several future growing seasons. Rapid and accurate detection of weed patches in field is the first step of site-specific weed management. In this study, object detection-based convolutional neural network models were trained and evaluated over low-altitude unmanned aerial vehicle (UAV) imagery for mid- to late-season weed detection in soybean fields. The performance of two object detection models, Faster RCNN and the Single Shot Detector (SSD), were evaluated and compared in terms of weed detection performance using mean Intersection over Union (IoU) and inference speed. It was found that the Faster RCNN model with 200 box proposals had similar good weed detection performance to the SSD model in terms of precision, recall, f1 score, and IoU, as well as a similar inference time. The precision, recall, f1 score and IoU were 0.65, 0.68, 0.66 and 0.85 for Faster RCNN with 200 proposals, and 0.66, 0.68, 0.67 and 0.84 for SSD, respectively. However, the optimal confidence threshold of the SSD model was found to be much lower than that of the Faster RCNN model, which indicated that SSD might have lower generalization performance than Faster RCNN for mid- to late-season weed detection in soybean fields using UAV imagery. The performance of the object detection model was also compared with patch-based CNN model. The Faster RCNN model yielded a better weed detection performance than the patch-based CNN with and without overlap. The inference time of Faster RCNN was similar to patch-based CNN without overlap, but significantly less than patch-based CNN with overlap. Hence, Faster RCNN was found to be the best model in terms of weed detection performance and inference time among the different models compared in this study. This work is important in understanding the potential and identifying the algorithms for an on-farm, near real-time weed detection and management.


IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 129300-129309 ◽  
Author(s):  
Wenchi Ma ◽  
Kaidong Li ◽  
Guanghui Wang

Author(s):  
Jie Xu

Abstract Recent advances in the field of object detection and face recognition have made it possible to develop practical video surveillance systems with embedded object detection and face recognition functionalities that are accurate and fast enough for commercial uses. In this paper, we compare some of the latest approaches to object detection and face recognition and provide reasons why they may or may not be amongst the best to be used in video surveillance applications in terms of both accuracy and speed. It is discovered that Faster R-CNN with Inception ResNet V2 is able to achieve some of the best accuracies while maintaining real-time rates. Single Shot Detector (SSD) with MobileNet, on the other hand, is incredibly fast and still accurate enough for most applications. As for face recognition, FaceNet with Multi-task Cascaded Convolutional Networks (MTCNN) achieves higher accuracy than advances such as DeepFace and DeepID2+ while being faster. An end-to-end video surveillance system is also proposed which could be used as a starting point for more complex systems. Various experiments have also been attempted on trained models with observations explained in detail. We finish by discussing video object detection and video salient object detection approaches which could potentially be used as future improvements to the proposed system.


Sign in / Sign up

Export Citation Format

Share Document