ECCV 2016 accepted papers on Architectures and efficiency
25 of the 415 papers accepted at ECCV 2016 (6.0%) are on Architectures and efficiency. They fall into 4 topics.
See ECCV 2016 in the interactive map
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 · 14 papers
- Identity Mappings in Deep Residual Networks (11,157 citations)
- XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks (4,554 citations)
- Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks (464 citations)
- Instance-Sensitive Fully Convolutional Networks (433 citations)
- Learning to Count with CNN Boosting (310 citations)
- Less Is More: Towards Compact CNNs (304 citations)
- LSTM-CF: Unifying Context Modeling and Fusion with LSTMs for RGB-D Scene Labeling (181 citations)
- Convolutional Oriented Boundaries (140 citations)
- Network of Experts for Large-Scale Image Categorization (100 citations)
- Top-Down Learning for Structured Labeling with Convolutional Pseudoprior (41 citations)
- Accelerating Convolutional Neural Networks with Dominant Convolutional Kernel and Knowledge Pre-regression (22 citations)
- Collaborative Layer-Wise Discriminative Learning in Deep Neural Networks (20 citations)
and 2 more in the interactive map.
RNNs and normalization · 6 papers
- Deep Networks with Stochastic Depth (2,247 citations)
- A Siamese Long Short-Term Memory Architecture for Human Re-identification (552 citations)
- Fast Optical Flow Using Dense Inverse Search (393 citations)
- Extending Long Short-Term Memory for Multi-View Structured Learning (141 citations)
- Learning Visual Storylines with Skipping Recurrent Neural Networks (35 citations)
- A Simple Hierarchical Pooling Data Structure for Loop Closure (14 citations)
Quantization · 3 papers
- Revisiting Additive Quantization (61 citations)
- Approximate Search with Quantized Sparse Representations (20 citations)
- Large-Scale R-CNN with Classifier Adaptive Quantization (6 citations)
Pruning and efficient tuning · 2 papers
- Learning Without Forgetting (4,525 citations)
- CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples (565 citations)
