Soft-NMS — Improving Object Detection with One Line of Code

Bodla, Navaneeth; Singh, Bharat; Chellappa, Rama; Davis, Larry S.

doi:10.1109/iccv.2017.593

Cited by 1,774 publications

(1,070 citation statements)

References 32 publications

Supporting

Mentioning

1,065

Contrasting

Unclassified

Order By: Relevance

“…Repulsion loss [1] is designed to directly penalize the predicted box for shifting to the other ground-truth objects or other predicted boxes associated with different ground truths. Soft NMS [23] decays the detection scores of boxes depending on the overlap and eliminates no boxes in the process to achieve high recall. Adaptive NMS [24] applies a dynamic suppression threshold according to the target density.…”

Section: Occlusion Handlingmentioning

confidence: 99%

Semantic head enhanced pedestrian detection in a crowd

Wang

2020

Neurocomputing

View full text Add to dashboard Cite

Pedestrian detection in the crowd is a challenging task because of intra-class occlusion. More prior information is needed for the detector to be robust against it. Human head area is naturally a strong cue because of its stable appearance, visibility and relative location to body. Inspired by it, we adopt an extra branch to conduct semantic head detection in parallel with traditional body branch. Instead of manually labeling the head regions, we use weak annotations inferred directly from body boxes, which is named as 'semantic head'. In this way, the head detection is formulated into using a special part of labeled box to detect the corresponding part of human body, which surprisingly improves the performance and robustness to occlusion. Moreover, the head-body alignment structure is explicitly explored by introducing Alignment Loss, which functions in a self-supervised manner. Based on these, we propose the head-body alignment net (HBAN) in this work, which aims to enhance pedestrian detection by fully utilizing the human head prior. Comprehensive evaluations are conducted to demonstrate the effectiveness of HBAN on CityPersons dataset.

show abstract

Section: Occlusion Handlingmentioning

confidence: 99%

Semantic head enhanced pedestrian detection in a crowd

Wang

2020

Neurocomputing

View full text Add to dashboard Cite

show abstract

“…After that, we apply non-maximum suppression (NMS) with IoU threshold 0.6. In NMS procedure, instead of removing adjacent bounding boxes, we divide their scores by 10, which is similar to Soft-NMS [4]. In Figure 5, performance of proposals generated by our method are quantified and compared with popular proposal techniques [18,26,38,2,7,5,9,21,1,30,31,17].…”

Section: Object Proposalsmentioning

confidence: 99%

Universal Bounding Box Regression and Its Applications

Lee¹,

Kwak²,

Cho³

2019

Computer Vision – ACCV 2018

View full text Add to dashboard Cite

Bounding-box regression is a popular technique to refine or predict localization boxes in recent object detection approaches. Typically, bounding-box regressors are trained to regress from either region proposals or fixed anchor boxes to nearby bounding boxes of a pre-defined target object classes. This paper investigates whether the technique is generalizable to unseen classes and is transferable to other tasks beyond supervised object detection. To this end, we propose a class-agnostic and anchor-free box regressor, dubbed Universal Bounding-Box Regressor (UBBR), which predicts a bounding box of the nearest object from any given box. Trained on a relatively small set of annotated images, UBBR successfully generalizes to unseen classes, and can be used to improve localization in many vision problems. We demonstrate its effectivenss on weakly supervised object detection and object discovery.

show abstract

“…We only use one scale, namely 512 × 512 for both training and testing. During inference, we run soft NMS [3] on the model outputs with the standard deviation parameter of 0.55 in the Gaussian weighting function.…”

Section: Implementation Details and Baselinesmentioning

confidence: 99%

Detecting 11K Classes: Large Scale Object Detection Without Fine-Grained Bounding Boxes

Yang

Chen

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Recent advances in deep learning greatly boost the performance of object detection. State-of-the-art methods such as Faster-RCNN, FPN and R-FCN have achieved high accuracy in challenging benchmark datasets. However, these methods require fully annotated object bounding boxes for training, which are incredibly hard to scale up due to the high annotation cost. Weakly-supervised methods, on the other hand, only require image-level labels for training, but the performance is far below their fully-supervised counterparts. In this paper, we propose a semi-supervised large scale fine-grained detection method, which only needs bounding box annotations of a smaller number of coarsegrained classes and image-level labels of large scale finegrained classes, and can detect all classes at nearly fullysupervised accuracy. We achieve this by utilizing the correlations between coarse-grained and fine-grained classes with shared backbone, soft-attention based proposal reranking, and a dual-level memory module. Experiment results show that our methods can achieve close accuracy on object detection to state-of-the-art fully-supervised methods on two large scale datasets, ImageNet and OpenImages, with only a small fraction of fully annotated classes.

show abstract

Soft-NMS — Improving Object Detection with One Line of Code

Cited by 1,774 publications

References 32 publications

Semantic head enhanced pedestrian detection in a crowd

Semantic head enhanced pedestrian detection in a crowd

Universal Bounding Box Regression and Its Applications

Detecting 11K Classes: Large Scale Object Detection Without Fine-Grained Bounding Boxes

Contact Info

Product

Resources

About