mlsanityData to 7 Oct 2026

Highly Efficient Salient Object Detection with 100K Parameters

arXiv 2003.05643 ↗First posted 12 Mar 2020cs.CVcited by 79

Not on the map. Why?

Abstract (arXiv)

Salient object detection models often demand a considerable amount of computation cost to make precise prediction for each pixel, making them hardly applicable on low-power devices. In this paper, we aim to relieve the contradiction between computation cost and model performance by improving the network efficiency to a higher degree. We propose a flexible convolutional module, namely generalized OctConv (gOctConv), to efficiently utilize both in-stage and cross-stages multi-scale features, while reducing the representation redundancy by a novel dynamic weight decay scheme. The effective dynamic weight decay scheme stably boosts the sparsity of parameters during training, supports learnable number of channels for each scale in gOctConv, allowing 80% of parameters reduce with negligible performance drop. Utilizing gOctConv, we build an extremely light-weighted model, namely CSNet, which achieves comparable performance with about 0.2% parameters (100k) of large models on popular salient object detection benchmarks.

Key facts. Cited by 79 papers in the map (data through 7 Oct 2026). First posted on arXiv 12 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 17; 2023: 18; 2024: 9; 2025: 6; 2026: 7.

79 papers.

Sort:

By area:

  1. 2021Concealed Object Detectioncited by 212
  2. 2021Simultaneously Localize, Segment and Rank the Camouflaged Objectscited by 198
  3. 2021Visual Saliency Transformercited by 175
  4. 2021Pixel Difference Networks for Efficient Edge Detectioncited by 167
  5. 2022Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detectioncited by 145
  6. 2021Salient Object Detection via Integrity Learningcited by 133
  7. 2021Uncertainty-aware Joint Salient Object and Camouflaged Object Detectioncited by 114
  8. 2021Large-scale Unsupervised Semantic Segmentationcited by 98
  9. 2023Feature Shrinkage Pyramid for Camouflaged Object Detection with Transformerscited by 87
  10. 2022High-resolution Iterative Feedback Network for Camouflaged Object Detectioncited by 78
  11. 2021Only Train Once: A One-Shot Neural Network Training And Pruning Frameworkcited by 70
  12. 2022Boundary-Guided Camouflaged Object Detectioncited by 68
38 more
  1. 2022Deep Gradient Learning for Efficient Camouflaged Object Detectioncited by 65
  2. 2023VSCode: General Visual Salient and Camouflaged Object Detection with 2D Prompt Learningcited by 58
  3. 2021Enhanced Boundary Learning for Glass-like Object Segmentationcited by 48
  4. 2022Weakly-Supervised Camouflaged Object Detection with Scribble Annotationscited by 43
  5. 2021Global2Local: Efficient Structure Search for Video Action Segmentationcited by 42
  6. 2021One-Shot Object Affordance Detection in the Wildcited by 40
  7. 2021Disentangled High Quality Salient Object Detectioncited by 35
  8. 2021Group Collaborative Learning for Co-Salient Object Detectioncited by 32
  9. 2021One-Shot Affordance Detectioncited by 32
  10. 2023OTOV2: Automatic, Generic, User-Friendlycited by 32
  11. 2021Densely Nested Top-Down Flows for Salient Object Detectioncited by 23
  12. 2021Salient Objects in Cluttercited by 19
  13. 2022Towards Deeper Understanding of Camouflaged Object Detectioncited by 19
  14. 2023Towards Diverse Binary Segmentation via A Simple yet General Gated Networkcited by 17
  15. 2023Salient Object Detection in Optical Remote Sensing Images Driven by Transformercited by 17
  16. 2023Lightweight Salient Object Detection in Optical Remote-Sensing Images via Semantic Matching and Edge Alignmentcited by 16
  17. 2022Lightweight Salient Object Detection in Optical Remote Sensing Images via Feature Correlationcited by 15
  18. 2021Learning to Detect Instance-level Salient Objects Using Complementary Image Labelscited by 14
  19. 2025Automatic Joint Structured Pruning and Quantization for Efficient Neural Network Training and Compressioncited by 14
  20. 2022MFFN: Multi-view Feature Fusion Network for Camouflaged Object Detectioncited by 12
  21. 2024MoExtend: Tuning New Experts for Modality and Task Extensioncited by 12
  22. 2022RF-Next: Efficient Receptive Field Search for Convolutional Neural Networkscited by 10
  23. 2022Panoramic Video Salient Object Detection with Ambisonic Audio Guidancecited by 9
  24. 2022CF-YOLO: Cross Fusion YOLO for Object Detection in Adverse Weather with a High-quality Real Snow Datasetcited by 7
  25. 2023Internal-External Boundary Attention Fusion for Glass Surface Segmentationcited by 6
  26. 2023OTOv3: Automatic Architecture-Agnostic Neural Network Training and Compression from Structured Pruning to Erasing Operatorscited by 6
  27. 2021Middle-level Fusion for Lightweight RGB-D Salient Object Detectioncited by 5
  28. 2022DFTR: Depth-supervised Fusion Transformer for Salient Object Detectioncited by 5
  29. 2022Towards Complex Backgrounds: A Unified Difference-Aware Decoder for Binary Segmentationcited by 5
  30. 2023Rethinking Lightweight Salient Object Detection via Network Depth-Width Tradeoffcited by 5
  31. 2024MINet: Multi-scale Interactive Network for Real-time Salient Object Detection of Strip Steel Surface Defectscited by 5
  32. 2021SHD360: A Benchmark Dataset for Salient Human Detection in 360deg Videoscited by 4
  33. 2022Pyramidal Attention for Saliency Detectioncited by 4
  34. 2022A lightweight multi-scale context network for salient object detection in optical remote sensing imagescited by 4
  35. 2022Single-Stage Open-world Instance Segmentation with Cross-task Consistency Regularizationcited by 4
  36. 2022Towards Stable Co-saliency Detection and Object Co-segmentationcited by 4
  37. 2023Camouflaged Object Detection with Feature Grafting and Distractor Awarecited by 4
  38. 2023M3Net: Multilevel, Mixed and Multistage Attention Network for Salient Object Detectioncited by 4

50 most cited of 79 All 79 (JSON)