The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TNNLS.2020.3028964 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Words before pictures: the role of language in biasing visual attention.
Front Psychol
December 2024
Department of Developmental Psychology and Socialization, University of Padova, Padua, Italy.
Background: The present study investigated whether semantic processing of word and object primes can bias visual attention using top-down influences, even within an exogenous cueing framework. We hypothesized that real words and familiar objects would more effectively bias attentional engagement and target detection than pseudowords or pseudo-objects, as they can trigger prior knowledge to influence attention orienting and target detection.
Methods: To examine this, we conducted two web-based eye-tracking experiments that ensured participants maintained central fixation on the screen during remote data collection.
An Unusual Case of Low Vision and Anisocoria Considered a Neurological Finding in the Emergency Department: Ocular Siderosis.
Turk J Ophthalmol
December 2024
Kastamonu Training and Research Hospital, Clinic of Ophthalmology, Kastamonu, Türkiye.
We present the case of a patient who came to the emergency department with a significant decrease in vision and dilated pupil in the left eye. Since neurological pathologies were primarily considered, diffusion brain magnetic resonance imaging (MRI) and brain computed tomography (CT) were requested. After the results were reported as normal, we were consulted.
View Article and Find Full Text PDFA Study on Real-time Detection of Rice Diseases in Farmlands Based on Multi-dimensional Data Fusion.
Plant Dis
January 2025
Biotechnology, plant protection, Nongsheng Group C735, Zijin Campus, Zhejiang University, Hangzhou, Zhejiang, China, 310058;
To meet the need of crop leaf disease detection in complex scenarios, this study designs a method based on the computing power of mobile devices that ensures both detection accuracy and real-time efficiency, offering significant practical application value. Based on a comparison with existing mainstream detection models, this paper proposes a target detection and recognition algorithm, TG_YOLOv5, which utilizes multi-dimensional data fusion on the YOLOv5 model. The triplet attention mechanism and C3CBAM module are incorporated into the network structure to capture connections between spatial and channel dimensions of input feature maps, thereby enhancing the model's feature extraction capabilities without significantly increasing the parameter count.
View Article and Find Full Text PDFVisual short-term memory binding and attentional processes during object-scene integration are preserved in mild cognitive impairment.
Cortex
December 2024
Department of Psychology, Sapienza University of Rome, Rome, Italy; IRCCS Fondazione Santa Lucia, Rome, Italy. Electronic address:
Binding, a critical cognitive process likely mediated by attention, is essential for creating coherent object representations within a scene. This process is vulnerable in individuals with dementia, who exhibit deficits in visual working memory (VWM) binding, primarily tested using abstract arrays of standalone objects. To explore how binding operates in more realistic settings across the lifespan, we examined the impact of object saliency and semantic consistency on VWM binding and the role of overt attention.
View Article and Find Full Text PDFEdge-guided feature fusion network for RGB-T salient object detection.
Front Neurorobot
December 2024
Department of Information Engineering, Shanghai Maritime University, Shanghai, China.
Introduction: RGB-T Salient Object Detection (SOD) aims to accurately segment salient regions in both visible light and thermal infrared images. However, many existing methods overlook the critical complementarity between these modalities, which can enhance detection accuracy.
Methods: We propose the Edge-Guided Feature Fusion Network (EGFF-Net), which consists of cross-modal feature extraction, edge-guided feature fusion, and salience map prediction.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!