A Novel Region of Interest Extraction Layer for Instance Segmentation

Rossi, Leonardo; Karimi, Akbar; Prati, Andrea

doi:10.1109/icpr48806.2021.9412258

Cited by 51 publications

(20 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, proposal region or sample assign strategy etc. [55][56][57]59] for detector were also designed to optimize the precision in many studies. Given the poor performance of small targets compared to large-scale targets, the optimization of target detectors has limitations for small-scale target performance improvement, and the results have been validated.…”

Section: Discussionmentioning

confidence: 99%

“…Input size #params CARAFE [54] (3, 600, 400) 46.73 M SABL [55] (3, 600, 400) 41.91 M Dynamic R-CNN [56] (3, 600, 400) 41.12 M PISA [57] (3, 600, 400) 41.12 M Mask R-CNN [58] (3, 600, 400) 43.75 M Groie [59] (3, 600, 400) 43.02 M Faster R-CNN [27] (3, 600, 400) 41.12 M BFP Net (3, 600, 400) 42.55 M 13 Plant Phenomics superior to the comparison models, as a whole, the BFP Net has a strong generalization ability for detecting the general object, especially for small target detection.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

Sun

Chen³

et al. 2022

Plant Phenomics

View full text Add to dashboard Cite

Despite of significant achievements made in the detection of target fruits, small fruit detection remains a great challenge, especially for immature small green fruits with a few pixels. The closeness of color between the fruit skin and the background greatly increases the difficulty of locating small target fruits in the natural orchard environment. In this paper, we propose a balanced feature pyramid network (BFP Net) for small apple detection. This network can balance information mapped to small apples from two perspectives: multiple-scale fruits on the different layers of FPN and a characteristic of a new extended feature from the output of ResNet50 conv1. Specifically, we design a weight-like feature fusion architecture on the lateral connection and top-down structure to alleviate the small-scale information imbalance on the different layers of FPN. Moreover, a new extended layer from ResNet50 conv1 is embedded into the lowest layer of standard FPN, and a decoupled-aggregated module is devised on this new extended layer of FPN to complement spatial location information and relieve the problem of locating small apple. In addition, a feature Kullback-Leibler distillation loss is introduced to transfer favorable knowledge from the teacher model to the student model. Experimental results show that APS of our method reaches 47.0%, 42.2%, and 35.6% on the benchmark of the GreenApple, MinneApple, and Pascal VOC, respectively. Overall, our method is not only slightly better than some state-of-the-art methods but also has a good generalization performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

Sun

Chen³

et al. 2022

Plant Phenomics

View full text Add to dashboard Cite

show abstract

“…YOLACT (Bolya et al, 2019) performs realtime instance segmentation by having two parallel pipelines, where one generates a set of prototype masks, while the other predicts the coefficients of masks per-instance. General ROI Extraction (Rossi et al, 2020) introduces non-local building blocks and attention mechanisms to attend to multiple layers of FPN, to extract a coherent subset of features for integrating in two stage methods. Prime Sample Attention (PISA) in object detection (Cao et al, 2020) assesses how different samples from the dataset con-tribute to the overall performance in mean AP.…”

Section: Instance Segmentation Methodsmentioning

confidence: 99%

The MIS Check-Dam Dataset for Object Detection and Instance Segmentation Tasks

Chintan¹,

Kumar²,

Damani³

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep learning has led to many recent advances in object detection and instance segmentation, among other computer vision tasks. These advancements have led to wide application of deep learning based methods and related methodologies in object detection tasks for satellite imagery. In this paper, we introduce MIS Check-Dam, a new dataset of check-dams from satellite imagery for building an automated system for the detection and mapping of check-dams, focusing on the importance of irrigation structures used for agriculture. We review some of the most recent object detection and instance segmentation methods and assess their performance on our new dataset. We evaluate several single stage, two-stage and attention based methods under various network configurations and backbone architectures. The dataset and the pre-trained models are available at https://www.cse.iitb.ac.in/gramdrishti/

show abstract

“…With the development of deep convolutional networks, many object detection frameworks have been proposed. Some Region-based Convolutional Neural Network (R-CNN) [1] frameworks, such as Fast R-CNN [3], Faster R-CNN [2], Libra R-CNN [23], Generic Region of Interest Extractor (GRoIE) [25], CBNet [27], ThunderNet [28] and CSPNet [29], fuse features from different levels to obtain single-level features that simultaneously include semantic information and location information. In its representative method the object region proposal component is achieved by selective search [4], and the classification component is obtained using a convolutional neural network (CNN).…”

Section: Introductionmentioning

confidence: 99%

BFF R-CNN: Balanced Feature Fusion for Object Detection

Liu

WANG

et al. 2022

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

In the neck part of a two-stage object detection network, feature fusion is generally carried out in either a top-down or bottom-up manner. However, two types of imbalance may exist: feature imbalance in the neck of the model and gradient imbalance in the region of interest extraction layer due to the scale changes of objects. The deeper the network is, the more abstract the learned features are, that is to say, more semantic information can be extracted. However, the extracted image background, spatial location, and other resolution information are less. In contrast, the shallow part can learn little semantic information, but a lot of spatial location information. We propose the Both Ends to Centre to Multiple Layers (BEtM) feature fusion method to solve the feature imbalance problem in the neck and a Multi-level Region of Interest Feature Extraction (MRoIE) layer to solve the gradient imbalance problem. In combination with the Regionbased Convolutional Neural Network (R-CNN) framework, our Balanced Feature Fusion (BFF) method offers significantly improved network performance compared with the Faster R-CNN architecture. On the MS COCO 2017 dataset, it achieves an average precision (AP) that is 1.9 points and 3.2 points higher than those of the Feature Pyramid Network (FPN) Faster R-CNN framework and the Generic Region of Interest Extractor (GRoIE) framework, respectively.

show abstract

A Novel Region of Interest Extraction Layer for Instance Segmentation

Cited by 51 publications

References 18 publications

BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

The MIS Check-Dam Dataset for Object Detection and Instance Segmentation Tasks

BFF R-CNN: Balanced Feature Fusion for Object Detection

Contact Info

Product

Resources

About