Realworld 3D Object Recognition Using a 3D Extension of the HOG Descriptor and a Depth Camera

Vilar, Cristian; Krug, Silvia; O’Nils, Mattias

doi:10.3390/s21030910

Cited by 8 publications

(8 citation statements)

References 31 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the above property, each lattice descriptor abstracts the shape variations with respect to diffeomorphism, while global isometry only influences the result to a limited degree; see Figure 3 for a visualization. Note that the local descriptor can easily be substituted by multiple species of descriptors; for instance, the 3D HoG descriptor [ 34 ] defines a process including an explicit conversion from raw point cloud data back to a depth map, then computing the statistics feature on a fixed angled plane to form the representation. However, this kind of operation inevitably loses the fineness of raw scan results, as the depth map yields a regular 2D domain.…”

Section: Methodsmentioning

confidence: 99%

Facial Expression Recognition with Geometric Scattering on 3D Point Clouds

Cheng

et al. 2022

Sensors

View full text Add to dashboard Cite

As one of the pioneering data representations, the point cloud has shown its straightforward capacity to depict fine geometry in many applications, including computer graphics, molecular structurology, modern sensing signal processing, and more. However, unlike computer graphs obtained with auxiliary regularization techniques or from syntheses, raw sensor/scanner (metric) data often contain natural random noise caused by multiple extrinsic factors, especially in the case of high-speed imaging scenarios. On the other hand, grid-like imaging techniques (e.g., RGB images or video frames) tend to entangle interesting aspects with environmental variations such as pose/illuminations with Euclidean sampling/processing pipelines. As one such typical problem, 3D Facial Expression Recognition (3D FER) has been developed into a new stage, with remaining difficulties involving the implementation of efficient feature abstraction methods for high dimensional observations and of stabilizing methods to obtain adequate robustness in cases of random exterior variations. In this paper, a localized and smoothed overlapping kernel is proposed to extract discriminative inherent geometric features. By association between the induced deformation stability and certain types of exterior perturbations through manifold scattering transform, we provide a novel framework that directly consumes point cloud coordinates for FER while requiring no predefined meshes or other features/signals. As a result, our compact framework achieves 78.33% accuracy on the Bosphorus dataset for expression recognition challenge and 77.55% on 3D-BUFE.

show abstract

Section: Methodsmentioning

confidence: 99%

Facial Expression Recognition with Geometric Scattering on 3D Point Clouds

Cheng

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The 3DHOG approach requires previous point cloud pre-processing to segment the objects from the unstructured 3D points. The pre-processing segmentation causes a higher computational cost but also leads to segmentation errors, which affect the recognition performance [5]. In Experiment 2, we used a new method of the object segmentation using the same labels we used for the YOLOv4-Tiny approach.…”

Section: Evaluation Of the Methodsmentioning

confidence: 99%

“…To compare this work to the results obtained in a previous 3DHOG evaluation [4,5], we use a synthetic dataset for Experiment 2 (Figure 3b) to train the SVM classifier when using the 3DHOG object descriptor. We extracted the synthetic feet data from the ModelNet40 dataset [16] using the objects of the class person and manually segmenting the feet portion of each representation (Figure 5).…”

Section: Synthetic Dataset Generationmentioning

confidence: 99%

“…Currently, there are two different approaches for object recognition: (1) hand-crafted features and (2) deep learning. In previous research, we evaluated the performance and limitations of a hand-crafted approach to object recognition based on an extension of the 3D data of a histogram of oriented gradients (3DHOG) object descriptor in combination with a Support Vector Machine (SVM) classifier [3][4][5]. The results showed relatively good recognition performance and good processing times, making it suitable for real-time operation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of 2D-/3D-Feet-Detection Methods for Semi-Autonomous Powered Wheelchair Navigation

et al. 2021

Self Cite

View full text Add to dashboard Cite

Powered wheelchairs have enhanced the mobility and quality of life of people with special needs. The next step in the development of powered wheelchairs is to incorporate sensors and electronic systems for new control applications and capabilities to improve their usability and the safety of their operation, such as obstacle avoidance or autonomous driving. However, autonomous powered wheelchairs require safe navigation in different environments and scenarios, making their development complex. In our research, we propose, instead, to develop contactless control for powered wheelchairs where the position of the caregiver is used as a control reference. Hence, we used a depth camera to recognize the caregiver and measure at the same time their relative distance from the powered wheelchair. In this paper, we compared two different approaches for real-time object recognition using a 3DHOG hand-crafted object descriptor based on a 3D extension of the histogram of oriented gradients (HOG) and a convolutional neural network based on YOLOv4-Tiny. To evaluate both approaches, we constructed Miun-Feet—a custom dataset of images of labeled caregiver’s feet in different scenarios, with backgrounds, objects, and lighting conditions. The experimental results showed that the YOLOv4-Tiny approach outperformed 3DHOG in all the analyzed cases. In addition, the results showed that the recognition accuracy was not improved using the depth channel, enabling the use of a monocular RGB camera only instead of a depth camera and reducing the computational cost and heat dissipation limitations. Hence, the paper proposes an additional method to compute the caregiver’s distance and angle from the Powered Wheelchair (PW) using only the RGB data. This work shows that it is feasible to use the location of the caregiver’s feet as a control signal for the control of a powered wheelchair and that it is possible to use a monocular RGB camera to compute their relative positions.

show abstract

“…According to the Equation ( 7), the centroid of M 1 is m 1 = Rm 0 . According to the Equation ( 8) and (10),…”

Section: ) Second-order Moments Of Solid Modelmentioning

confidence: 99%

A Pose-Normalization Method for Casting Voxel Models Using Second-Order Central Moment Matrix

Guo

Dong

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Within the framework of casting process design, the efficient retrieval of three-dimensional (3D) computer-aided design (CAD) models could result in significant time and cost savings. However, this technique still suffers from inefficiency and inaccuracy because of the wide variety of casting models' poses. In this study, a method for normalizing the poses of casting models is proposed. This method constructs the transformation matrix through the eigendecomposition of the second-order central moment matrix calculated from the voxel casting model. Then the transformation matrix is applied to the casting model to get a normalized pose. An assessment approach for pose normalization is also suggested in the study, which measures the distance between poses normalized based on multiple poses of the same model. The study demonstrates that the pose-normalization approach reliably transforms distinct poses of the same model into a unified pose. The mean distance between normalized poses is 0.016 and the minimum distance is 0.010. The method's effects improve when the voxel size reduces.

show abstract

Realworld 3D Object Recognition Using a 3D Extension of the HOG Descriptor and a Depth Camera

Cited by 8 publications

References 31 publications

Facial Expression Recognition with Geometric Scattering on 3D Point Clouds

Facial Expression Recognition with Geometric Scattering on 3D Point Clouds

Evaluation of 2D-/3D-Feet-Detection Methods for Semi-Autonomous Powered Wheelchair Navigation

A Pose-Normalization Method for Casting Voxel Models Using Second-Order Central Moment Matrix

Contact Info

Product

Resources

About