mlsanityData to 14 Oct 2026

STEERER: Resolving Scale Variations for Counting and Localization via Selective Inheritance Learning

arXiv 2308.10468 ↗First posted 21 Aug 2023cs.CVcited by 41

Not on the map. Why?

Abstract (arXiv)

Scale variation is a deep-rooted problem in object counting, which has not been effectively addressed by existing scale-aware algorithms. An important factor is that they typically involve cooperative learning across multi-resolutions, which could be suboptimal for learning the most discriminative features from each scale. In this paper, we propose a novel method termed STEERER (SelecTivE inhERitance lEaRning) that addresses the issue of scale variations in object counting. STEERER selects the most suitable scale for patch objects to boost feature extraction and only inherits discriminative features from lower to higher resolution progressively. The main insights of STEERER are a dedicated Feature Selection and Inheritance Adaptor (FSIA), which selectively forwards scale-customized features at each scale, and a Masked Selection and Inheritance Loss (MSIL) that helps to achieve high-quality density maps across all scales. Our experimental results on nine datasets with counting and localization tasks demonstrate the unprecedented scale generalization ability of STEERER. Code is available at urlhttps://github.com/taohan10200/STEERER.

Key facts. Cited by 41 papers in the map (data through 14 Oct 2026). Builds on 54 of its 85 references that are in the map. First posted on arXiv 21 Aug 2023.

Builds on

54 of 85 references.

References · 54

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  5. 2015Learning Deep Features for Discriminative Localizationcited by 2,436
  6. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
  7. 2018Path Aggregation Network for Instance Segmentationcited by 798
  8. 2018CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenescited by 267
  9. 2018Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowdscited by 186
  10. 2019Bayesian Loss for Crowd Count Estimation with Point Supervisioncited by 162
  11. 2021Rethinking Counting and Localization in Crowds:A Purely Point-Based Frameworkcited by 152
  12. 2020NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localizationcited by 151
38 more
  1. 2018Context-Aware Crowd Countingcited by 150
  2. 2019Learning from Synthetic Data for Crowd Counting in the Wildcited by 148
  3. 2020Distribution Matching for Crowd Countingcited by 143
  4. 2020JHU-CROWD++: Large-Scale Crowd Counting Dataset and A Benchmark Methodcited by 109
  5. 2017Switching Convolutional Neural Network for Crowd Countingcited by 93
  6. 2016A Large Contextual Dataset for Classification, Detection and Counting of Cars with Deep Learningcited by 86
  7. 2019Locate, Size and Count: Accurately Resolving People in Dense Crowds via Detectioncited by 73
  8. 2022Boosting Crowd Counting via Multifaceted Attentioncited by 73
  9. 2016Finding Tiny Facescited by 72
  10. 2020Localization in the Crowd with Topological Constraintscited by 68
  11. 2017Generating High-Quality Crowd Density Maps using Contextual Pyramid CNNscited by 55
  12. 2018Where are the Blobs: Counting by Localization with Point Supervisioncited by 54
  13. 2022Rethinking Spatial Invariance of Convolutional Networks for Object Countingcited by 54
  14. 2019From Open Set to Closed Set: Counting Objects by Spatial Divide-and-Conquercited by 48
  15. 2019Crowd Counting with Deep Structured Scale Integration Networkcited by 46
  16. 2019Multi-Level Bottom-Top and Top-Bottom Feature Fusion for Crowd Countingcited by 36
  17. 2019Learning Spatial Awareness to Improve Crowd Countingcited by 36
  18. 2020A Self-Training Approach for Point-Supervised Object Detection and Counting in Crowdscited by 36
  19. 2021Focal Inverse Distance Transform Maps for Crowd Localizationcited by 36
  20. 2019Learn to Scale: Generating Multipolar Normalized Density Maps for Crowd Countingcited by 32
  21. 2019Locality-constrained Spatial Transformer Network for Video Crowd Countingcited by 30
  22. 2017A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimationcited by 29
  23. 2020Adaptive Mixture Regression Network with Local Counting Map for Crowd Countingcited by 28
  24. 2017FCN-rLSTM: Deep Spatio-Temporal Neural Networks for Vehicle Counting in City Camerascited by 27
  25. 2017Multi-scale Convolutional Neural Networks for Crowd Countingcited by 26
  26. 2017TasselNet: Counting maize tassels in the wild via local counts regression networkcited by 26
  27. 2017People, Penguins and Petri Dishes: Adapting Object Counting Models To New Visual Domains And Object Types Without Forgettingcited by 25
  28. 2020Learning Independent Instance Maps for Crowd Localizationcited by 24
  29. 2017Crowd counting via scale-adaptive convolutional neural networkcited by 23
  30. 2019Crowd Counting Using Scale-Aware Attention Networkscited by 20
  31. 2020Neuron Linear Transformation: Modeling the Domain Shift for Crowd Countingcited by 19
  32. 2019C^3 Framework: An Open-source PyTorch Code for Crowd Countingcited by 18
  33. 2022DR.VIC: Decomposition and Reasoning for Video Individual Countingcited by 18
  34. 2020ZoomCount: A Zooming Mechanism for Crowd Counting in Static Imagescited by 17
  35. 2022Redesigning Multi-Scale Neural Network for Crowd Countingcited by 17
  36. 2020Focus on Semantic Consistency for Cross-domain Crowd Understandingcited by 16
  37. 2022Revisiting Crowd Counting: State-of-the-art, Trends, and Future Perspectivescited by 16
  38. 2018Crowd Counting by Adaptively Fusing Predictions from an Image Pyramidcited by 13

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 13; 2025: 15; 2026: 10.

41 papers.

Sort:

By area:

  1. 2024A Survey on Data Augmentation in Large Model Eracited by 31
  2. 2024Gramformer: Learning Crowd Counting via Graph-Modulated Transformercited by 19
  3. 2023Regressor-Segmenter Mutual Prompt Learning for Crowd Countingcited by 16
  4. 2024OmniCount: Multi-label Object Counting with Semantic-Geometric Priorscited by 16
  5. 2024CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classificationcited by 12
  6. 2023Semi-Supervised Crowd Counting with Contextual Modeling: Facilitating Holistic Understanding of Crowd Scenescited by 6
  7. 2025Point-to-Region Loss for Semi-Supervised Point-Based Crowd Countingcited by 6
  8. 2024Diffusion-based Data Augmentation for Object Counting Problemscited by 5
  9. 2025Taste More, Taste Better: Diverse Data and Strong Model Boost Semi-Supervised Crowd Countingcited by 5
  10. 2024Robust Zero-Shot Crowd Counting and Localization With Adaptive Resolution SAMcited by 4
  11. 2025Video Individual Counting for Moving Dronescited by 4
  12. 2024The Effectiveness of a Simplified Model Structure for Crowd Countingcited by 3
29 more
  1. 2026UNICBench: UNIfied Counting Benchmark for MLLMcited by 3
  2. 2023Shifted Autoencoders for Point Annotation Restoration in Object Countingcited by 2
  3. 2024Few-shot Object Localizationcited by 2
  4. 2024Dynamic Proxy Domain Generalizes the Crowd Localization by Better Binary Segmentationcited by 2
  5. 2025ZIP: Scalable Crowd Counting via Zero-Inflated Poisson Modelingcited by 2
  6. 2024Enhancing Zero-shot Counting via Language-guided Exemplar Learningcited by 1
  7. 2024CrowdMAC: Masked Crowd Density Completion for Robust Crowd Density Forecastingcited by 1
  8. 2025Depth-Assisted Network for Indiscernible Marine Object Counting with Adaptive Motion-Differentiated Feature Encodingcited by 1
  9. 2025Consistent-Point: Consistent Pseudo-Points for Semi-Supervised Crowd Counting and Localizationcited by 1
  10. 2025Exploring Scale Shift in Crowd Localization under the Context of Domain Generalizationcited by 1
  11. 2025Decoupling What to Count and Where to See for Referring Expression Countingcited by 1
  12. 2026Video Individual Counting and Tracking from Moving Drones: A Benchmark and Methodscited by 1
  13. 2024CountMamba: Exploring Multi-directional Selective State-Space Models for Plant Countingcited by 0
  14. 2024Text-guided Zero-Shot Object Localizationcited by 0
  15. 2025ANTHROPOS-V: benchmarking the novel task of Crowd Volume Estimationcited by 0
  16. 2025EHNet: An Efficient Hybrid Network for Crowd Counting and Localizationcited by 0
  17. 2025Crowd Detection Using Very-Fine-Resolution Satellite Imagerycited by 0
  18. 2025Towards Conditional Feature Alignment for Cross-Domain Countingcited by 0
  19. 2025Local Information Matters: A Rethink of Crowd Countingcited by 0
  20. 2025Count2Density: Crowd Density Estimation without Location-level Annotationscited by 0
  21. 2025Active View Selection for Scene-level Multi-view Crowd Counting and Localization with Limited Labeling Budgetcited by 0
  22. 2026RepSFNet : A Single Fusion Network with Structural Reparameterization for Crowd Countingcited by 0
  23. 2026FSCA-Net: Feature-Separated Cross-Attention Network for Robust Multi-Dataset Trainingcited by 0
  24. 2026SynMVCrowd: A Large Synthetic Benchmark for Multi-view Crowd Counting and Localizationcited by 0
  25. 2026DenseControl: Instance-Level Controllable Synthesis of Dense Crowd Imagecited by 0
  26. 2026Multi-Modal Hyper-Graph Fusion for Low-Light Crowd Countingcited by 0
  27. 2026DCA-MoE: Spatially Adaptive Cross-Layer Fusion and Density-Routed Experts for Crowd Countingcited by 0
  28. 2026Object Counting Across Modalities: Taxonomies, Benchmarks, Applications, and Open Challengescited by 0
  29. 2026CrowdCue: Specialist-Cue Conditioning for Vision-Language Crowd Countingcited by 0