Soft-NMS -- Improving Object Detection With One Line of Code

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

doi:10.48550/arxiv.1704.04503

Cited by 46 publications

(53 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NMS is performed on a per-class basis using IoU thresholds 0.5 and 0.75. We also performed testing with soft-NMS [17] with very minor code changes, and slightly improved the model's mAP. The inference performance is compared to the results from a previous research paper [16].…”

Section: Testing and Experimental Resultsmentioning

confidence: 99%

Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series

Zhu¹,

Sundaresan²,

Rupe³

2020

2020 IEEE International Conference on Prognostics and Health Management (ICPHM)

View full text Add to dashboard Cite

Proactive network maintenance (PNM) is the concept of using data from a network to identify and locate network faults, many or all of which could worsen to become service failures. The separation between the network fault and the service failure affords early detection of problems in the network to allow PNM to take place. Consequently, PNM is a form of prognostics and health management (PHM).The problem of localizing and classifying anomalies on 1dimensional data series has been under research for years. We introduce a new algorithm that leverages Deep Convolutional Neural Networks to efficiently and accurately detect anomalies and events on data series, and it reaches 97.82% mean average precision (mAP) in our evaluation.

show abstract

Section: Testing and Experimental Resultsmentioning

confidence: 99%

Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series

Zhu¹,

Sundaresan²,

Rupe³

2020

2020 IEEE International Conference on Prognostics and Health Management (ICPHM)

View full text Add to dashboard Cite

show abstract

“…With the above process, our BSN++ can generate the proposal candidates set as Ψ p = {ϕ n = (t s , t e , p ϕ )} Np n=1 , where N p is the number of proposals. Since the generated proposals may overlap with each other to various degrees, we conduct Soft-NMS [1] algorithm to suppress the confidence scores of redundant proposals. Final, the proposals set is Ψ p = {ϕ n = (t s , t e , p ϕ )} Np n=1 , where p ϕ is the decayed score of proposal ϕ n .…”

Section: Inferencementioning

confidence: 99%

“…Akin to object proposals for object detection in images, temporal action proposal indicates the temporal intervals containing the actions and plays an important role in temporal action detection. It has been commonly recognized that high-quality proposals usually have two crucial properties: (1) the generated proposals should cover the action instances temporally with both high recall and temporal overlapping; (2) the quality of proposals should be evaluated comprehensively and accurately, thus providing a overall confidence for later retrieving step.…”

Section: Introductionmentioning

confidence: 99%

“…[20] is a typical method in this type which proposes the Boundary-Sensitive Network (BSN) to generate proposals with flexible durations and reliable confidence scores. Though BSN achieves convincing performance, it still suffers from three main drawbacks: (1) only the local details around the boundaries have been employed in BSN to predict boundaries, without taking advantage of the rich contexts through the whole video sequence; (2) the proposalproposal relations fail to be considered for confidence evaluation; (3) the imbalance data distribution between positive/negative proposals and temporal durations is neglected.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Complementary Boundary Generator with Scale-Invariant Relation Modeling for Temporal Action Localization: Submission to ActivityNet Challenge 2020

Su,

Feng,

Shao

et al. 2020

Preprint

View full text Add to dashboard Cite

This technical report presents an overview of our solution used in the submission to ActivityNet Challenge 2020 Task 1 (temporal action localization/detection). Temporal action localization requires to not only precisely locate the temporal boundaries of action instances, but also accurately classify the untrimmed videos into specific categories. In this paper, we decouple the temporal action localization task into two stages (i.e. proposal generation and classification) and enrich the proposal diversity through exhaustively exploring the influences of multiple components from different but complementary perspectives. Specifically, in order to generate high-quality proposals, we consider several factors including the video feature encoder, the proposal generator, the proposal-proposal relations, the scale imbalance, and ensemble strategy. Finally, in order to obtain accurate detections, we need to further train an optimal video classifier to recognize the generated proposals. Our proposed scheme achieves the state-of-the-art performance on the temporal action localization task with 42.26 average mAP on the challenge testing set.

show abstract

“…For task 1, through merging the improved-BSN & APN model and CAR model by Soft-NMS [1], we achieve 69.85 AUC score on both validation set and testing server.…”

Section: Temporal Action Proposal and Localizationmentioning

confidence: 99%

Multi-Granularity Fusion Network for Proposal and Activity Localization: Submission to ActivityNet Challenge 2019 Task 1 and Task 2

Su,

Zhao,

Liu

2019

Preprint

View full text Add to dashboard Cite

This technical report presents an overview of our solution used in the submission to ActivityNet Challenge 2019 Task 1 (temporal action proposal generation) and Task 2 (temporal action localization/detection). Temporal action proposal indicates the temporal intervals containing the actions and plays an important role in temporal action localization. Top-down and bottom-up methods are the two main categories used for proposal generation in the existing literature. In this paper, we devise a novel Multi-Granularity Fusion Network (MGFN) to combine the proposals generated from different frameworks for complementary filtering and confidence re-ranking. Specifically, we consider the diversity comprehensively from multiple perspectives, e.g. the characteristic aspect, the data aspect, the model aspect and the result aspect. Our MGFN achieves the state-of-theart performance on the temporal action proposal task with 69.85 AUC score and the temporal action localization task with 38.90 mAP on the challenge testing set.

show abstract

Soft-NMS -- Improving Object Detection With One Line of Code

Cited by 46 publications

References 0 publications

Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series

Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series

Complementary Boundary Generator with Scale-Invariant Relation Modeling for Temporal Action Localization: Submission to ActivityNet Challenge 2020

Multi-Granularity Fusion Network for Proposal and Activity Localization: Submission to ActivityNet Challenge 2019 Task 1 and Task 2

Contact Info

Product

Resources

About