mlsanityData to 14 Oct 2026

Dynamic Computational Time for Visual Attention

arXiv 1703.10332 ↗First posted 30 Mar 2017cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

We propose a dynamic computational time model to accelerate the average processing time for recurrent visual attention (RAM). Rather than attention with a fixed number of steps for each input image, the model learns to decide when to stop on the fly. To achieve this, we add an additional continue/stop action per time step to RAM and use reinforcement learning to learn both the optimal attention policy and stopping policy. The modification is simple but could dramatically save the average computational time while keeping the same recognition performance as RAM. Experimental results on CUB-200-2011 and Stanford Cars dataset demonstrate the dynamic computational model can work effectively for fine-grained image recognition.The source code of this paper can be obtained from https://github.com/baidu-research/DT-RAM

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). First posted on arXiv 30 Mar 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 9; 2022: 4; 2023: 6; 2024: 0; 2025: 3; 2026: 1.

23 papers.

Sort:

By area:

  1. 2021Dynamic Neural Networks: A Surveycited by 408
  2. 2021MOOD: Multi-level Out-of-distribution Detectioncited by 94
  3. 2021Differentiable Patch Selection for Image Recognitioncited by 68
  4. 2023Variational Information Pursuit for Interpretable Predictionscited by 25
  5. 2021Human Attention in Fine-grained Classificationcited by 21
  6. 2023Internally Rewarded Reinforcement Learningcited by 14
  7. 2023Computation-efficient Deep Learning for Computer Vision: A Surveycited by 14
  8. 2021A Probabilistic Hard Attention Model For Sequentially Observed Scenescited by 12
  9. 2022SALISA: Saliency-based Input Sampling for Efficient Video Object Detectioncited by 12
  10. 2023BirdSAT: Cross-View Contrastive Masked Autoencoders for Bird Species Classification and Mappingcited by 12
  11. 2021Re-rank Coarse Classification with Local Region Enhanced Features for Fine-Grained Image Recognitioncited by 11
  12. 2022Fine-Grained Visual Classification using Self Assessment Classifiercited by 8
11 more
  1. 2022Boosted Dynamic Neural Networkscited by 8
  2. 2023Foveation in the Era of Deep Learningcited by 8
  3. 2021Learning Canonical 3D Object Representation for Fine-Grained Recognitioncited by 7
  4. 2021Progressive Co-Attention Network for Fine-grained Visual Classificationcited by 5
  5. 2025Locality-aware Concept Bottleneck Modelcited by 4
  6. 2023SlowFormer: Universal Adversarial Patch for Attack on Compute and Energy Efficiency of Inference Efficient Vision Transformerscited by 3
  7. 2025Emergence of Fixational and Saccadic Movements in a Multi-Level Recurrent Attention Model for Visioncited by 3
  8. 2021Sparsely ensembled convolutional neural network classifiers via reinforcement learningcited by 2
  9. 2022E^2TAD: An Energy-Efficient Tracking-based Action Detectorcited by 1
  10. 2025EnGraf-Net: Multiple Granularity Branch Network with Fine-Coarse Graft Grained for Classification Taskcited by 0
  11. 2026Exploring Budgeted Image Classification with Content-Sensitive Resource Allocationcited by 0