ICCV 2017 accepted papers on Architectures and efficiency
29 of the 621 papers accepted at ICCV 2017 (4.7%) are on Architectures and efficiency. They fall into 5 topics.
See ICCV 2017 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
CNN design · 17 papers
- Learning Efficient Convolutional Networks Through Network Slimming (2,860 citations)
- Generating High-Quality Crowd Density Maps Using Contextual Pyramid CNNs (731 citations)
- SSH: Single Stage Headless Face Detector (516 citations)
- Interleaved Group Convolutions (390 citations)
- Fast Image Processing With Fully-Convolutional Networks (363 citations)
- Is Second-Order Information Helpful for Large-Scale Visual Recognition? (301 citations)
- Rotation Equivariant Vector Field Networks (251 citations)
- Scaling the Scattering Transform: Deep Hybrid Networks (172 citations)
- Factorized Bilinear Models for Image Recognition (114 citations)
- Non-Linear Convolution Filters for CNN-Based Learning (97 citations)
- DualNet: Learn Complementary Features for Image Recognition (86 citations)
- Learning Bag-Of-Features Pooling for Deep Convolutional Neural Networks (79 citations)
and 5 more in the interactive map.
RNNs and normalization · 6 papers
- SubUNets: End-To-End Hand Shape and Continuous Sign Language Recognition (382 citations)
- Curriculum Dropout (107 citations)
- Learning Long-Term Dependencies for Action Recognition With a Biologically-Inspired Deep Network (72 citations)
- Centered Weight Normalization in Accelerating Training of Deep Neural Networks (71 citations)
- Unrolled Memory Inner-Products: An Abstract GPU Operator for Efficient Vision-Related Computations (5 citations)
- Truncating Wide Networks Using Binary Tree Architectures (4 citations)
Pruning and efficient tuning · 3 papers
- Channel Pruning for Accelerating Very Deep Neural Networks (2,753 citations)
- ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression (1,935 citations)
- Coordinating Filters for Faster Deep Neural Networks (128 citations)
Quantization · 2 papers
Neural architecture search · 1 paper
- Genetic CNN (930 citations)
