computer-vision

Dense Object Detection

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in dense-object-detection

Trend

Dataset

Best Model

Actions

SKU-110K

Libraries

i

Use these libraries to find dense-object-detection models and implementations

open-mmlab/mmdetection

4 papers 30,832

Datasets

DOTA

SKU110K

xView3-SAR

Subtasks

No subtasks available.

Most implemented papers

Precise Detection in Densely Packed Scenes

Aviv Eisenschtat, Eran Goldman, Roei Herzig, Oria Ratzon, Itsik Levi, J. Goldberger, Tal Hassner•Sun Mar 31 2019

This work proposes a novel, deep-learning based method for precise object detection, designed for such challenging settings as packed retail environments, and shows the method to outperform existing state-of-the-art with substantial margins.

211

Content

PaddlePaddle/PaddleDetection

4 papers 13,323

0

Paper Graph

PP-YOLOE: An evolved version of YOLO

Qingqing Dang, Baohua Lai, Shangliang Xu, Xinxin Wang, Wenyu Lv, Qinyao Chang, Cheng Cui, Kaipeng Deng, Guanzhong Wang, Shengyun Wei, Yuning Du•Tue Mar 29 2022

PP-YOLOE, an industrial state-of-the-art object detector with high performance and friendly deployment is presented, using anchor-free paradigm, more powerful backbone and neck equipped with CSPRepResStage, ET-head and dynamic label assignment algorithm TAL.

385 0

Paper Graph

Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection

Wenhai Wang, Xiang Li, Xiaolin Hu, Jian Yang, Jinhui Tang, Jun Li•Tue Nov 24 2020

This paper explores a completely novel and different perspective to perform LQE – based on the learned distributions of the four parameters of the bounding box – and develops a considerably lightweight Distribution-Guided Quality Predictor (DGQP) for reliable LqE based on GFLV1, thus producing GFLv2.

294 0

Paper Graph

Soft Anchor-Point Object Detection

M. Savvides, Chenchen Zhu, Zhiqiang Shen, Fangyi Chen•Tue Nov 26 2019

This work boosts the performance of the anchor-point detector over the key-point counterparts while maintaining the speed advantage, and proposes a simple yet effective training strategy with soft-weighted anchor points and soft-selected pyramid levels to address the false attention issue within each pyramid level and the feature selection issue across all the pyramid levels.

159 0

Paper Graph

A Solution to Product detection in Densely Packed Scenes

Tianze Rong, Yanjia Zhu, Yichao Xiong, Hongxiang Cai•Wed Jul 22 2020

This work proposed a random crop strategy to ensure both the sampling rate and input scale is relatively sufficient as a contrast to the regular random crop and adopted some of trick and optimized the hyper-parameters.

9 0

Paper Graph

AutoAssign: Differentiable Label Assignment for Dense Object Detection

Zeming Li, Zhengkai Jiang, Songtao Liu, Jianfeng Wang, Benjin Zhu, Fuhang Zong•Mon Jul 06 2020

An anchor-free object detector with a fully differentiable label assignment strategy, named AutoAssign, that automatically determines positive/negative samples by generating positive and negative weight maps to modify each location's prediction dynamically.

248 0

Paper Graph

BorderDet: Border Feature for Dense Object Detection

Zeming Li, Songtao Liu, Han Qiu, Yuchen Ma•Mon Jul 20 2020

This paper designs a novel detection architecture called BorderDet, which explicitly exploits the border information for stronger classification and more accurate localization and proposes a simple and efficient operator called Border-Align to extract "border features" from the extreme point of the border to enhance the point feature.

145 0

Paper Graph

Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection

Wenhai Wang, Xiang Li, Xiaolin Hu, Jian Yang, Jinhui Tang, Jun Li, Shuo Chen, Lijun Wu•Sun Jun 07 2020

Improved representations of quality estimation into the class prediction vector to form a joint representation of localization quality and classification, and use a vector to represent arbitrary distribution of box locations are designed.

1625 0

Paper Graph

Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement

Xiuquan Hou, Senlin Zhang, Ping Wei, Mei-qin Liu, Badong Chen•Sat Mar 23 2024

This paper demonstrates that suboptimal two-stage selection strategies result in scale bias and redundancy due to the mismatch between selected queries and objects in two-stage initial-ization, and proposes hierarchical salience filtering refinement, which performs transformer encoding only on filtered discriminative queries, for a bet-ter trade-off between computational efficiency and precision.

81 0

Paper Graph

Adding a benchmark result helps the community track progress.