Grid R-CNN

By Xin Lu, Buyu Li, Yuxin Yue, Quanquan Li, Junjie Yan. 2018

495 citations recorded by the source corpus.

This paper proposes a novel object detection framework named Grid R-CNN, which adopts a grid guided localization mechanism for accurate object detection. Different from the traditional regression based methods, the Grid R-CNN captures the spatial information explicitly and enjoys the position sensitive property of fully convolutional architecture. Instead of using only two independent points, we design a multi-point supervision formulation to encode more clues in order to reduce the impact of inaccurate prediction of specific points. To take the full advantage of the correlation of points in a grid, we propose a two-stage information fusion strategy to fuse feature maps of neighbor grid points. The grid guided localization approach is easy to be extended to different state-of-the-art detection frameworks. Grid R-CNN leads to high quality object localization, and experiments demonstrate that it achieves a 4.1% AP gain at IoU=0.8 and a 10.0% AP gain at IoU=0.9 on COCO benchmark compared to Faster R-CNN with Res50 backbone and FPN architecture.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

1. Xiaogang Wang
2. Stephen Lin
3. Cheng Chi
4. Zheng Zhang
5. Han Hu
6. Qi Tian
7. Lingxi Xie
8. Fangyun Wei

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Xiaogang Wang
3. Dacheng Tao
4. Ming-Hsuan Yang
5. Yi Yang
6. Ling Shao
7. Chunhua Shen
8. Qi Tian

Citations in the corpus

• Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (2019, methodology): “The second stage of Grid R-CNN [39] locates objects via predicting grid points with the position sensitive merits of FCN and then determining the bounding box guided by the grid.”
• MMDetection: Open MMLab Detection Toolbox and Benchmark (2019, neutral): “• Grid R-CNN [20]: a grid guided localization mechanism as an alternative to bounding box regression, proposed in 2018.”
• CenterNet: Keypoint Triplets for Object Detection (2019, background): “4] addresses the problem of overfitting at training and quality mismatch at inference by training a sequence of detectors with increasing IoU thresholds. The keypoint-based object detection approaches [40, 28] are proposed to avoid the disadvantages of using anchor boxes and bounding boxes regression”
• Conditional DETR for Fast Training Convergence (2021, methodology): “Typical methods include YOLOv1 [30], CornerNet [21], ExtremeNet [50], CenterNet [49, 6], FCOS [39] and others [23, 28, 52, 19, 51, 22, 15, 46, 47].”
• RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder (2020, background): “Some approaches [30, 15, 16, 7, 21, 39, 26] first detect these individual points and then compose bounding boxes from them.”
• Revisiting the Sibling Head in Object Detector (2020, methodology): “Based on Faster R-CNN, we directly conduct several experiments with different backbones and Table.5 summarizes the detailed performance on COCO minival.”
• Tiny Object Detection in Aerial Images (2021, methodology): “Anchor-free Detector: For keypoint-based detectors, they first locate several pre-defined or self-learned keypoints, and then generate bounding boxes to detect objects [22], such as CornerNet [28], Grid R-CNN [29], CenterNet [30], and RepPoints [31].”
• RepPoints V2: Verification Meets Regression for Object Detection (2020, background): “Hybrid approaches Most detectors are hybrid, for example, those built on anchors or proposals [6, 5, 21, 16, 1, 19].”
• Recent Advances in Deep Learning for Object Detection (2019, methodology): “Some other efforts [150, 151] focus on designing a uni-fied framework with modified objective functions.”
• D2Det: Towards High Quality Object Detection and Instance Segmentation (2020, methodology): “3) our dense local regression (DLR) with the keypoint-based localization utilized in the recently introduced Grid R-CNN [36] and its variant Grid R-CNN Plus [37].”
• Corner Proposal Network for Anchor-free, Two-stage Object Detection (2020, unknown): “6 Grid R-CNN [28] X-101-64x4d 800 43.”
• CentripetalNet: Pursuing High-Quality Keypoint Pairs for Object Detection (2020, unknown): “Other works cooperating with anchor-based detectors are proposed to deal with different issues, such as improving anchor selection procedure [33], refining feature learning process [39, 18], optimizing location prediction method [24], and improving the loss function [30, 16].”
• GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving (2019, background): “nerate 3D candidate boxes with a much smaller number but even higher accuracy. 1https://drive.google.com/file/d/188BxA_ jlhHHpxCXk3SxPBA5qkmk53PIt/view?usp=sharing Since state-of-the-art 2D detectors [20, 17, 13, 16, 15] can provide reliable 2D bounding boxes for objects, several works use 2D box as”
• Localization Distillation for Object Detection (2022, background): “Object localization is a fundamental issue in object detection [18], [33], [44], [64], [65], [73], [74], [75], [80], [93].”
• Detecting tiny objects in aerial images: A normalized Wasserstein distance and a new benchmark (2022, unknown): “…2020) ResNet-50-FPN 11.3 28.1 7.4 1.4 8.6 17.8 32.2 21.8 FCOS (Tian et al., 2019) ResNet-50-FPN 12.0 30.2 7.3 2.2 11.1 16.6 26.9 26.0 Grid R-CNN (Lu et al., 2019) ResNet-50-FPN 14.3 31.1 11.0 0.1 11.0 25.7 36.7 23.5 ours: RetinaNet w/ NWD-RKA ResNet-50-FPN 10.5 28.5 5.2 3.5 11.0 15.2 24.2 25.7…”