Deep Learning for Computer Vision

  1. Modern CNN Architectures
    1. Efficient Architectures
      1. MobileNet v1
        1. Depthwise Separable Convolutions
          1. Width Multiplier
            1. Resolution Multiplier
            2. MobileNet v2
              1. Inverted Residuals
                1. Linear Bottlenecks
                2. MobileNet v3
                  1. Neural Architecture Search
                    1. Squeeze-and-Excitation
                    2. EfficientNet
                      1. Compound Scaling
                        1. Model Scaling Dimensions
                          1. EfficientNet Family
                          2. ShuffleNet
                            1. Channel Shuffle Operation
                              1. Group Convolutions
                            2. Attention-based Architectures
                              1. Squeeze-and-Excitation Networks
                                1. Channel Attention
                                  1. Global Information
                                  2. Convolutional Block Attention Module
                                    1. Channel and Spatial Attention
                                    2. Non-local Networks
                                      1. Self-attention in CNNs
                                        1. Long-range Dependencies
                                      2. Neural Architecture Search
                                        1. Search Space Design
                                          1. Search Strategy
                                            1. Performance Estimation
                                              1. DARTS
                                              2. ProxylessNAS
                                                1. Direct Hardware Optimization

                                            Previous

                                            4. Classical CNN Architectures

                                            Go to top

                                            Next

                                            6. Core Computer Vision Tasks