Learning to Separate: Detecting Heavily-Occluded Objects in Urban Scenes

Yang, Chenhongyi; Ablavsky, Vitaly; Wang, Kaihong; Feng, Qianjin; Betke, Margrit

doi:10.48550/arxiv.1912.01674

Cited by 3 publications

(3 citation statements)

References 43 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al [114] applied a dynamic suppression threshold to an instance based on the target density. Yang et al [219] developed bounding-boxlevel Semantics-Geometry Embedding (SGE) to distinguish two heavily-overlapping boxes by combining detection results. Huang et al [81] proposed R 2 NMS, which uses the IoU between the visible regions to determine whether or not the two full-body boxes overlap.…”

Section: Pure Cnn Based Pedestrian Detection Methodsmentioning

confidence: 99%

“…Adaptive NMS [114] uses an adaptive threshold according to the density of pedestrians. SGE [219] uses a dynamic threshold of NMS for crowd pedestrian detection. CrowdDetection [34] uses one proposal to predict multiple instances, and employs a set NMS to remove duplicate bounding-boxes after checking whether two boundingboxes come from the same proposal.…”

Section: Occlusionmentioning

confidence: 99%

“…The learning-based schemes learn a mapping to retain the most accurate bounding-boxes. The heuristic-based schemes include the greedy NMS, Soft-NMS [7], SGE-NMS [219], and Adaptive NMS [114]. The learning-based methods include Gnet [73] and Relation Network [76].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

From Handcrafted to Deep Features for Pedestrian Detection: A Survey

Cao

Pang

Xie

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Pedestrian detection is an important but challenging problem in computer vision, especially in human-centric tasks. Over the past decade, significant improvement has been witnessed with the help of handcrafted features and deep features. Here we present a comprehensive survey on recent advances in pedestrian detection. First, we provide a detailed review of single-spectral pedestrian detection that includes handcrafted features based methods and deep features based approaches. For handcrafted features based methods, we present an extensive review of approaches and find that handcrafted features with large freedom degrees in shape and space have better performance. In the case of deep features based approaches, we split them into pure CNN based methods and those employing both handcrafted and CNN based features. We give the statistical analysis and tendency of these methods, where feature enhanced, part-aware, and post-processing methods have attracted main attention. In addition to single-spectral pedestrian detection, we also review multi-spectral pedestrian detection, which provides more robust features for illumination variance. Furthermore, we introduce some related datasets and evaluation metrics, and compare some representative methods. We conclude this survey by emphasizing open problems that need to be addressed and highlighting various future directions. Researchers can track an up-to-date list at https://github.com/JialeCao001/PedSurvey.

show abstract

Section: Pure Cnn Based Pedestrian Detection Methodsmentioning

confidence: 99%

Section: Occlusionmentioning

confidence: 99%

See 1 more Smart Citation

From Handcrafted to Deep Features for Pedestrian Detection: A Survey

Cao

Pang

Xie

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

Ref-NMS: Breaking Proposal Bottlenecks in Two-Stage Referring Expression Grounding

Chen¹,

Ma²,

Xiao³

et al. 2020

Preprint

View full text Add to dashboard Cite

The prevailing framework for solving referring expression grounding is based on a two-stage process: 1) detecting proposals with an object detector and 2) grounding the referent to one of the proposals. Existing two-stage solutions mostly focus on the grounding step, which aims to align the expressions with the proposals. In this paper, we argue that these methods overlook an obvious mismatch between the roles of proposals in the two stages: they generate proposals solely based on the detection confidence (i.e., expression-agnostic), hoping that the proposals contain all right instances in the expression (i.e., expression-aware). Due to this mismatch, current two-stage methods suffer from a severe performance drop between detected and ground-truth proposals. To this end, we propose Ref-NMS, which is the first method to yield expression-aware proposals at the first stage. Ref-NMS regards all nouns in the expression as critical objects, and introduces a lightweight module to predict a score for aligning each box with a critical object. These scores can guide the NMS operation to filter out the boxes irrelevant to the expression, increasing the recall of critical objects, resulting in a significantly improved grounding performance. Since Ref-NMS is agnostic to the grounding step, it can be easily integrated into any state-of-the-art two-stage method. Extensive ablation studies on several backbones, benchmarks, and tasks consistently demonstrate the superiority of Ref-NMS.

show abstract

VL-NMS: Breaking Proposal Bottlenecks in Two-Stage Visual-Language Matching

Chen

Zhang

et al. 2021

Preprint

View full text Add to dashboard Cite

The prevailing framework for matching multimodal inputs is based on a two-stage process: 1) detecting proposals with an object detector and 2) matching text queries with proposals. Existing two-stage solutions mostly focus on the matching step. In this paper, we argue that these methods overlook an obvious mismatch between the roles of proposals in the two stages: they generate proposals solely based on the detection confidence (i.e., query-agnostic), hoping that the proposals contain all instances mentioned in the text query (i.e., query-aware). Due to this mismatch, chances are that proposals relevant to the text query are suppressed during the filtering process, which in turn bounds the matching performance. To this end, we propose VL-NMS, which is the first method to yield query-aware proposals at the first stage. VL-NMS regards all mentioned instances as critical objects, and introduces a lightweight module to predict a score for aligning each proposal with a critical object. These scores can guide the NMS operation to filter out proposals irrelevant to the text query, increasing the recall of critical objects, resulting in a significantly improved matching performance. Since VL-NMS is agnostic to the matching step, it can be easily integrated into any state-of-the-art two-stage matching methods. We validate the effectiveness of VL-NMS on two multimodal matching tasks, namely referring expression grounding and image-text matching. Extensive ablation studies on several baselines and benchmarks consistently demonstrate the superiority of VL-NMS.

show abstract

Learning to Separate: Detecting Heavily-Occluded Objects in Urban Scenes

Cited by 3 publications

References 43 publications

From Handcrafted to Deep Features for Pedestrian Detection: A Survey

From Handcrafted to Deep Features for Pedestrian Detection: A Survey

Ref-NMS: Breaking Proposal Bottlenecks in Two-Stage Referring Expression Grounding

VL-NMS: Breaking Proposal Bottlenecks in Two-Stage Visual-Language Matching

Contact Info

Product

Resources

About