A survey on instance segmentation: state of the art

Hafiz, Abdul Mueed; Bhat, G. Mohiuddin

doi:10.1007/s13735-020-00195-x

Cited by 352 publications

(148 citation statements)

References 105 publications

Supporting

Mentioning

144

Contrasting

Unclassified

Order By: Relevance

“…Since the inception of this project in 2018, a variety of new instance segmentation frameworks have been proposed as surveyed by Hafiz and Bhat (2020) . For example, Mask Scoring RCNN that also makes segmentation based on detection of region proposals extends further with the inclusion of mask overlap scores and has slightly surpassed Mask R-CNN to achieve 39.6% AP on the same datasets ( Huang et al, 2019 ).…”

Section: Resultsmentioning

confidence: 99%

“…This poses challenges as most current algorithms cannot handle this type of task. Recent advances in Convolutional Neural Networks led to a variety of frameworks that can be used to perform instance segmentation on different levels ( Hafiz and Bhat, 2020 ). One of the most successful and popular approach was the Mask R-CNN framework which efficiently detects the object while simultaneously generating a high quality segmentation mask for each instance ( He et al, 2017 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Instance Segmentation to Estimate Consumption of Corn Ears by Wild Animals for GMO Preference Tests

Adke

Mogel²,

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

The Genetically Modified (GMO) Corn Experiment was performed to test the hypothesis that wild animals prefer Non-GMO corn and avoid eating GMO corn, which resulted in the collection of complex image data of consumed corn ears. This study develops a deep learning-based image processing pipeline that aims to estimate the consumption of corn by identifying corn and its bare cob from these images, which will aid in testing the hypothesis in the GMO Corn Experiment. Ablation uses mask regional convolutional neural network (Mask R-CNN) for instance segmentation. Based on image data annotation, two approaches for segmentation were discussed: identifying whole corn ears and bare cob parts with and without corn kernels. The Mask R-CNN model was trained for both approaches and segmentation results were compared. Out of the two, the latter approach, i.e., without the kernel, was chosen to estimate the corn consumption because of its superior segmentation performance and estimation accuracy. Ablation experiments were performed with the latter approach to obtain the best model with the available data. The estimation results of these models were included and compared with manually labeled test data with R2 = 0.99 which showed that use of the Mask R-CNN model to estimate corn consumption provides highly accurate results, thus, allowing it to be used further on all collected data and help test the hypothesis of the GMO Corn Experiment. These approaches may also be applied to other plant phenotyping tasks (e.g., yield estimation and plant stress quantification) that require instance segmentation.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Instance Segmentation to Estimate Consumption of Corn Ears by Wild Animals for GMO Preference Tests

Adke

Mogel²,

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…This relatively newer model configuration called top-down/bottom-up model has been used in attention-based object segmentation [110] and also in instance segmentation [73,31,29]. The architecture of the model is shown in Figure 2(d), wherein segmentation feature maps are first obtained by common bottom-up convolution techniques, and next a top-down refinement is done for fusing the data from deep to shallow layers into the mask.…”

Section: Bottom-up/top-down Modelmentioning

confidence: 99%

Attention mechanisms and deep learning for machine vision: A survey of the state of the art

Hafiz

Parah

Bhat

2021

Preprint

Self Cite

View full text Add to dashboard Cite

With the advent of state of the art nature-inspired pure attention based models i.e. transformers, and their success in natural language processing (NLP), their extension to machine vision (MV) tasks was inevitable and much felt. Subsequently, vision transformers (ViTs) were introduced which are giving quite a challenge to the established deep learning based machine vision techniques. However, pure attention based models/architectures like transformers require huge data, large training times and large computational resources. Some recent works suggest that combinations of these two varied fields can prove to build systems which have the advantages of both these fields. Accordingly, this state of the art survey paper is introduced which hopefully will help readers get useful information about this interesting and potential research area. A gentle introduction to attention mechanisms is given, followed by a discussion of the popular attention based deep architectures. Subsequently, the major categories of the intersection of attention mechanisms and deep learning for machine vision (MV) based are discussed. Afterwards, the major algorithms, issues and trends within the scope of the paper are discussed.

show abstract

“…Image segmentation is the pixel-based sorting task. There are two primary image segmentation forms: semantic segmentation (14) and instance segmentation (15). In semantic segmentation, each pixel of the image will be classified as belonging to a class; regardless of whether there is more than one object with the same class, the semantic segmentation will not distinguish between the objects as distinct entities.…”

Section: Pooling Layermentioning

confidence: 99%

Pre-Harvest Fruit Image Processing: A Brief Review

Gremes¹,

Krummenauer²,

Andrade³

et al. 2021

BJEDIS

View full text Add to dashboard Cite

Agriculture is essential for the development of human civilization. Methods that can precisely estimate the yield of a crop or to perform the harvest automatically using robots can decrease the costs involved and increase production efficiency. With the advancement of agriculture 4.0, current technologies like the internet of things, big data, and artificial intelligence have become more and more common. Systems that use image processing with Deep Learning methods are becoming viable in solving agricultural problems. Deep Learning is part of a large family of methods based on artificial neural networks that can mimic the human brain's work in data processing and pattern recognition for decision-making. Indeed, applications of Deep Learning techniques in agriculture are relatively recent. However, with the rapid advance in Deep Learning and its successful application in agriculture, many articles have been published in recent years. Thus, the main objective of this work was to carry out a brief bibliographic review of pre-harvest fruit image processing techniques, emphasizing the most recent applications using Deep Learning. As seen in the literature, Deep Learning is a promising tool for various agricultural activities, including fruit counting and automatic fruit harvesting using robots.

show abstract

A survey on instance segmentation: state of the art

Cited by 352 publications

References 105 publications

Instance Segmentation to Estimate Consumption of Corn Ears by Wild Animals for GMO Preference Tests

Instance Segmentation to Estimate Consumption of Corn Ears by Wild Animals for GMO Preference Tests

Attention mechanisms and deep learning for machine vision: A survey of the state of the art

Pre-Harvest Fruit Image Processing: A Brief Review

Contact Info

Product

Resources

About