mlsanityData to 7 Oct 2026

Accelerating Very Deep Convolutional Networks for Classification and Detection

arXiv 1505.06798 ↗First posted 26 May 2015cs.CV cs.LG cs.NEcited by 109

Not on the map. Why?

Abstract (arXiv)

This paper aims to accelerate the test-time computation of convolutional neural networks (CNNs), especially very deep CNNs that have substantially impacted the computer vision community. Unlike previous methods that are designed for approximating linear filters or linear responses, our method takes the nonlinear units into account. We develop an effective solution to the resulting nonlinear optimization problem without the need of stochastic gradient descent (SGD). More importantly, while previous methods mainly focus on optimizing one or two layers, our nonlinear method enables an asymmetric reconstruction that reduces the rapidly accumulated error when multiple (e.g., >=10) layers are approximated. For the widely used very deep VGG-16 model, our method achieves a whole-model speedup of 4x with merely a 0.3% increase of top-5 error in ImageNet classification. Our 4x accelerated VGG-16 model also shows a graceful accuracy degradation for object detection when plugged into the Fast R-CNN detector.

Key facts. Cited by 109 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 May 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 30; 2022: 30; 2023: 15; 2024: 16; 2025: 14; 2026: 4.

109 papers.

Sort:

By area:

  1. 2022Language model compression with weighted low-rank factorizationcited by 192
  2. 2021The Efficiency Misnomercited by 98
  3. 2024GPTVQ: The Blessing of Dimensionality for LLM Quantizationcited by 78
  4. 2022A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classificationcited by 64
  5. 2021Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimizationcited by 57
  6. 2022Revisiting Random Channel Pruning for Neural Network Compressioncited by 50
  7. 2023Pruning vs Quantization: Which is Better?cited by 45
  8. 2021A Dynamic Spatial-temporal Attention Network for Early Anticipation of Traffic Accidentscited by 42
  9. 2022NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Resultscited by 40
  10. 2022Compute-Efficient Deep Learning: Algorithmic Trends and Opportunitiescited by 38
  11. 2021BSQ: Exploring Bit-Level Sparsity for Mixed-Precision Neural Network Quantizationcited by 37
  12. 2021Towards Compact CNNs via Collaborative Compressioncited by 36
38 more
  1. 2024AlphaPruning: Using Heavy-Tailed Self Regularization Theory for Improved Layer-wise Pruning of Large Language Modelscited by 35
  2. 2021Compressing Neural Networks: Towards Determining the Optimal Layer-wise Decompositioncited by 33
  3. 2021Skip-Convolutions for Efficient Video Processingcited by 32
  4. 2022Enable Deep Learning on Mobile Devices: Methods, Systems, and Applicationscited by 32
  5. 2022Deep Networks on Toroids: Removing Symmetries Reveals the Structure of Flat Regions in the Landscape Geometrycited by 27
  6. 2022Data-Efficient Structured Pruning via Submodular Optimizationcited by 22
  7. 2025QVGen: Pushing the Limit of Quantized Video Generative Modelscited by 20
  8. 2021Double Similarity Distillation for Semantic Image Segmentationcited by 19
  9. 2021Collaborative Intelligence: Challenges and Opportunitiescited by 18
  10. 2021Towards Efficient Graph Convolutional Networks for Point Cloud Handlingcited by 18
  11. 2021Towards Accurate and Compact Architectures via Neural Architecture Transformercited by 17
  12. 2022LATITUDE: Robotic Global Localization with Truncated Dynamic Low-pass Filter in City-scale NeRFcited by 17
  13. 2024OSSCAR: One-Shot Structured Pruning in Vision and Language Models with Combinatorial Optimizationcited by 17
  14. 2022Compression-aware Training of Neural Networks using Frank-Wolfecited by 16
  15. 2021Pixel-by-Pixel Cross-Domain Alignment for Few-Shot Semantic Segmentationcited by 15
  16. 2022Towards Lightweight Black-Box Attacks against Deep Neural Networkscited by 13
  17. 2021Breast Cancer Diagnosis in Two-View Mammography Using End-to-End Trained EfficientNet-Based Convolutional Networkcited by 12
  18. 2022ZippyPoint: Fast Interest Point Detection, Description, and Matching through Mixed Precision Discretizationcited by 10
  19. 2022Delta Distillation for Efficient Video Processingcited by 10
  20. 2021Accelerating Neural ODEs Using Model Order Reductioncited by 9
  21. 2021Real-time Adversarial Perturbations against Deep Reinforcement Learning Policies: Attacks and Defensescited by 9
  22. 2021Manipulating Identical Filter Redundancy for Efficient Pruning on Deep and Complicated CNNcited by 9
  23. 2021Model Preserving Compression for Neural Networkscited by 9
  24. 2023PELA: Learning Parameter-Efficient Models with Low-Rank Approximationcited by 9
  25. 2023Regroup Median Loss for Combating Label Noisecited by 8
  26. 2022Confidence estimation of classification based on the distribution of the neural network output layercited by 7
  27. 2021Learning on Hardware: A Tutorial on Neural Network Accelerators and Co-Processorscited by 6
  28. 2024EPSD: Early Pruning with Self-Distillation for Efficient Model Compressioncited by 6
  29. 2024Activation Map Compression through Tensor Decomposition for Deep Learningcited by 6
  30. 2025LaX: Boosting Low-Rank Training of Foundation Models via Latent Crossingcited by 6
  31. 2023Low Rank Optimization for Efficient Deep Learning: Making A Balance between Compact Architecture and Fast Trainingcited by 5
  32. 2022RIBAC: Towards Robust and Imperceptible Backdoor Attack against Compact DNNcited by 4
  33. 2022CSTAR: Towards Compact and STructured Deep Neural Networks with Adversarial Robustnesscited by 4
  34. 2023Data-Free Dynamic Compression of CNNs for Tractable Efficiencycited by 4
  35. 2021Semi-tensor Product-based TensorDecomposition for Neural Network Compressioncited by 3
  36. 2022Nonlinear Initialization Methods for Low-Rank Neural Networkscited by 3
  37. 2022How to Train Unstable Looped Tensor Networkcited by 3
  38. 2022Training Quantised Neural Networks with STE Variants: the Additive Noise Annealing Algorithmcited by 3

50 most cited of 109 All 109 (JSON)