CentroidNetV2: A hybrid deep neural network for small-object segmentation and counting

Dijkstra, Klaas; Loosdrecht, Jaap van de; Atsma, Waatze A.; Schomaker, Lambert; Wiering, Marco A.

doi:10.1016/j.neucom.2020.10.075

Cited by 16 publications

(8 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During testing, the HS cube was tiled using a fixed grid of overlapping tiles and later recombined by only using the non-overlapping central part of each tile. This tiling with overlap helps to prevent border effects caused by the footprint of the segmentation models; a similar method was used by [32].…”

Section: Experiments and Resultsmentioning

confidence: 99%

On the Effect of Pre-Processing and Model Complexity for Plastic Analysis Using Short-Wave-Infrared Hyper-Spectral Imaging

Dijkstra¹,

Aghaei²,

Jaarsma³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The importance of plastic waste recycling is undeniable. In this respect, computer vision and deep learning enable solutions through the automated analysis of short-wave-infrared hyper-spectral images of plastics. In this paper, we offer an exhaustive empirical study to show the importance of efficient model selection for resolving the task of hyper-spectral image segmentation of various plastic flakes using deep learning. We assess the complexity level of generic and specialized models and infer their performance capacity: generic models are often unnecessarily complex. We introduce two variants of a specialized hyper-spectral architecture, PlasticNet, that outperforms several well-known segmentation architectures in both performance as well as computational complexity. In addition, we shed lights on the significance of signal pre-processing within the realm of hyper-spectral imaging. To complete our contribution, we introduce the largest, most versatile hyper-spectral dataset of plastic flakes of four primary polymer types.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

On the Effect of Pre-Processing and Model Complexity for Plastic Analysis Using Short-Wave-Infrared Hyper-Spectral Imaging

Dijkstra¹,

Aghaei²,

Jaarsma³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…While this representation allows to determine object centroids and consequently, enables locating and counting instances, it does not provide an instance-wise segmentation of the input. To this end, the authors of Dijkstra et al [13] enrich the centroid representation by an additional pixelwise representation of vectors pointing to the nearest object boundary. Similarly, the authors of Xie et al [14] propose a polar mask as instance representation, which defines each instance by the centre point of the object.…”

Section: Instance Representationmentioning

confidence: 99%

ConsInstancy: learning instance representations for semi-supervised panoptic segmentation of concrete aggregate particles

Coenen

Schack

Beyer

et al. 2022

Machine Vision and Applications

View full text Add to dashboard Cite

We present a semi-supervised method for panoptic segmentation based on ConsInstancy regularisation, a novel strategy for semi-supervised learning. It leverages completely unlabelled data by enforcing consistency between predicted instance representations and semantic segmentations during training in order to improve the segmentation performance. To this end, we also propose new types of instance representations that can be predicted by one simple forward path through a fully convolutional network (FCN), delivering a convenient and simple-to-train framework for panoptic segmentation. More specifically, we propose the prediction of a three-dimensional instance orientation map as intermediate representation and two complementary distance transform maps as final representation, providing unique instance representations for a panoptic segmentation. We test our method on two challenging data sets of both, hardened and fresh concrete, the latter being proposed by the authors in this paper demonstrating the effectiveness of our approach, outperforming the results achieved by state-of-the-art methods for semi-supervised segmentation. In particular, we are able to show that by leveraging completely unlabelled data in our semi-supervised approach the achieved overall accuracy (OA) is increased by up to 5% compared to an entirely supervised training using only labelled data. Furthermore, we exceed the OA achieved by state-of-the-art semi-supervised methods by up to 1.5%.

show abstract

“…While this representation allows to determine object centroids and consequently, enables locating and counting instances, it does not provide an instance-wise segmentation of the input. To this end, the authors of [13] enrich the centroid representation by an additional pixel-wise representation of vectors pointing to the nearest object boundary. Similarly, the authors of [14] propose a polar mask as instance representation, which defines each instance by the centre point of the object.…”

Section: Instance Representationmentioning

confidence: 99%

ConsInstancy: Learning Instance Representations for Semi-Supervised Panoptic Segmentation of Concrete Aggregate Particles

Coenen,

Schack,

Beyer

et al. 2022

Preprint

View full text Add to dashboard Cite

We present a semi-supervised method for panoptic segmentation based on ConsInstancy regularisation, a novel strategy for semi-supervised learning. It leverages completely unlabelled data by enforcing consistency between predicted instance representations and semantic segmentations during training in order to improve the segmentation performance. To this end, we also propose new types of instance representations that can be predicted by one simple forward path through a fully convolutional network (FCN), delivering a convenient and simple-to-train framework for panoptic segmentation. More specifically, we propose the prediction of a three-dimensional instance orientation map as intermediate representation and two complementary distance transform maps as final representation, providing unique instance representations for a panoptic segmentation. We test our method on two challenging data sets of both, hardened and fresh concrete, the latter being proposed by the authors in this paper demonstrating the effectiveness of our approach, outperforming the results achieved by state-of-the-art methods for semi-supervised segmentation. In particular, we are able to show that by leveraging completely unlabeled data in our semi-supervised approach the achieved overall accuracy (OA) is increased by up to 5% compared to an entirely supervised training using only labeled data. Furthermore, we exceed the OA achieved by state-of-the-art semi-supervised methods by up to 1.5%.

show abstract

CentroidNetV2: A hybrid deep neural network for small-object segmentation and counting

Abstract: Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Cited by 16 publications

References 34 publications

On the Effect of Pre-Processing and Model Complexity for Plastic Analysis Using Short-Wave-Infrared Hyper-Spectral Imaging

On the Effect of Pre-Processing and Model Complexity for Plastic Analysis Using Short-Wave-Infrared Hyper-Spectral Imaging

ConsInstancy: learning instance representations for semi-supervised panoptic segmentation of concrete aggregate particles

ConsInstancy: Learning Instance Representations for Semi-Supervised Panoptic Segmentation of Concrete Aggregate Particles

Contact Info

Product

Resources

About