Histogram of distances for local surface description

Kechagias‐Stamatis, Odysseas; Aouf, Nabil

doi:10.1109/icra.2016.7487402

Cited by 16 publications

(20 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The tuned parameters of each detector (Table 1) and descriptor ( Table 2) were used to maximize performance. The un-tuned parameters were fixed either to those proposed by the original authors or to their PCL implementation [6,38,50]. For the tuning process, we used the Oakland dataset and confirmed that SHOT has a stable description performance regardless of the description radius, whereas TriSI, FPFH and RoPS gain peak performance and then drop [4].…”

Section: Methodsmentioning

confidence: 99%

Performance evaluation of single and cross‐dimensional feature detection and description

2020

Self Cite

View full text Add to dashboard Cite

Three-dimensional local feature detection and description techniques are widely used for object registration and recognition applications. Although several evaluations of 3D local feature detection and description methods have already been published, these are constrained in a single dimensional scheme, i.e. either 3D or 2D methods that are applied onto multiple projections of the 3D data. However, cross-dimensional (mixed 2D and 3D) feature detection and description has yet to be investigated. Here, we evaluated the performance of both single and cross-dimensional feature detection and description methods on several 3D datasets and demonstrated the superiority of cross-dimensional over single-dimensional schemes. 2D / 3D Keypoint Detection and Feature Description Methods Keypoint DetectorsKeypoint detectors analyse the structure around a vertex or a pixel depending on the data domain (3D or 2D,

show abstract

Section: Methodsmentioning

confidence: 99%

Performance evaluation of single and cross‐dimensional feature detection and description

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…During trials we challenge B-HoD against current 3D pattern recognition algorithms presented in Table I. Specifically, we compare and contrast B-HoD with RoPS [6], SHOT [24], FPFH [10], 3DSC [8], USC [27], HoD [14], a binary version of HOD exploiting the quantization pipeline of [13] in combination with the subsampling of the currently proposed B-HoD descriptor. For better readability, this variant of HoD is notated as HoD (*) throughout this paper.…”

Section: A Experimental Setupmentioning

confidence: 99%

“…Although current 3D descriptors are a few and perform well [3]- [14], their computational and memory requirements may exceed the capabilities of a lightweight platform. A solution meeting those requirements can be exploiting a binary descriptor instead of a floating point as this allows a faster feature matching process along with a smaller descriptor footprint.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

B-HoD: A lightweight and fast binary descriptor for 3D object recognition and registration

Kechagias‐Stamatis

Aouf

Chermak

2017

2017 IEEE 14th International Conference on Networking, Sensing and Control (ICNSC)

Self Cite

View full text Add to dashboard Cite

3D object recognition and registration in computer vision applications has lately drawn much attention as it is capable of superior performance compared to its 2D counterpart. Although a number of high performing solutions do exist, it is still challenging to further reduce processing time and memory requirements to meet the needs of time critical applications. In this paper we propose an extension of the 3D descriptor Histogram of Distances (HoD) into the binary domain named the Binary-HoD (B-HoD). Our binary quantization procedure along with the proposed preprocessing step reduce an order of magnitude both processing time and memory requirements compared to current state of the art 3D descriptors. Evaluation on two popular low quality datasets shows its promising performance.

show abstract

“…Indicatively, features (data) extracted from the 3D domain are less affected by illumination variation and target pose changes [1] and describe the underlying structure of an object. With respect to future LIDAR based missiles, 3D ATR can improve weapon effectiveness against camouflage, concealment and deception techniques.…”

Section: Introductionmentioning

confidence: 99%

Evaluating 3D local descriptors for future LIDAR missiles with automatic target recognition capabilities

Kechagias‐Stamatis

Aouf

2017

The Imaging Science Journal

Self Cite

View full text Add to dashboard Cite

Evaluating 3D Local Descriptors for Future LIDAR Missiles with Automatic Target Recognition Capabilities Future Light Detection and Ranging seeker missiles incorporating 3D Automatic Target Recognition (ATR) capabilities can improve the missile's effectiveness in complex battlefield environments. Considering the progress of local 3D descriptors in the computer vision domain, this paper evaluates a number of these on highly credible simulated air-to-ground missile engagement scenarios. The latter take into account numerous parameters that have not been investigated yet by the literature including variable missile-target range, 6-Degrees-of-Freedom missile motion and atmospheric disturbances. Additionally, the evaluation process utilizes our suggested 3D ATR architecture that compared to current pipelines involves more post-processing layers aiming at further enhancing 3D ATR performance. Our trials reveal that computer vision algorithms are appealing for missile oriented 3D ATR.

show abstract

Histogram of distances for local surface description

Cited by 16 publications

References 37 publications

Performance evaluation of single and cross‐dimensional feature detection and description

Performance evaluation of single and cross‐dimensional feature detection and description

B-HoD: A lightweight and fast binary descriptor for 3D object recognition and registration

Evaluating 3D local descriptors for future LIDAR missiles with automatic target recognition capabilities

Contact Info

Product

Resources

About