Train 3D ConvNet

最新推荐文章于 2024-10-07 16:36:30 发布

paul_eric

最新推荐文章于 2024-10-07 16:36:30 发布

阅读量1.1k

点赞数

分类专栏： caffe 学习文章标签： caffe

本文链接：https://blog.csdn.net/ERIC_WHY/article/details/51445809

版权

caffe 同时被 2 个专栏收录

2 篇文章 0 订阅

订阅专栏

学习

1 篇文章 0 订阅

订阅专栏

本文详细介绍了如何使用3D卷积网络来学习视频中的时空特征，并通过创建体积均值、训练UCF101数据集以及测试网络性能的脚本来实现这一目标。该方法涉及网络结构设计、参数配置和关键步骤执行，旨在提高视频分类任务的准确性。

摘要由CSDN通过智能技术生成

Learning Spatiotemporal Features with 3D Convolutional Networks-2015-ICCV

1. create_volume_mean.sh

GLOG_logtostderr=1 ../../build/tools/compute_volume_mean_from_list.bin ../c3d_finetuning/train_01.lst 16 128 171 1 ucf101_train_mean.binaryproto 10

2. train_ucf101.sh

GLOG_logtostderr=1 ../../build/tools/train_net.bin conv3d_ucf101_solver.prototxt

3. test_ucf101.sh

GLOG_logtostderr=1 ../../build/tools/test_net.bin conv3d_ucf101_test.prototxt conv3d_ucf101_iter_60000 1396 GPU 0

a. conv3d_ucf101_solver.prototxt

train_net: "conv3d_ucf101_train.prototxt"
test_net: "conv3d_ucf101_test.prototxt"
test_iter: 100
test_interval: 1000
base_lr: 0.003
momentum: 0.9
weight_decay: 0.005
lr_policy: "step"
gamma: 0.1
stepsize: 20000
# Display every 20 iterations
display: 20
# The maximum number of iterations
max_iter: 60000
# snapshot intermediate results
snapshot: 1000
snapshot_prefix: "conv3d_ucf101"
# solver mode: CPU or GPU
solver_mode: GPU
device_id: 0

b. conv3d_ucf101_train.prototxt

name: "deep_c3d_ucf101"
layers {
  name: "data"
  type: VIDEO_DATA
  top: "data"
  top: "label"
  image_data_param {
    source: "../c3d_finetuning/train_01.lst"
    use_image: true
    mean_file: "ucf101_train_mean.binaryproto"
    batch_size: 30
    crop_size: 112
    mirror: true
    show_data: 0
    new_height: 128
    new_width: 171
    new_length: 16
    shuffle: true
  }
}
# ----------- 1st layer group ---------------
layers {
  name: "conv1a"
  type: CONVOLUTION3D
  bottom: "data"
  top: "conv1a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 64
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layers {
  name: "relu1a"
  type: RELU
  bottom: "conv1a"
  top: "conv1a"
}
layers {
  name: "pool1"
  type: POOLING3D
  bottom: "conv1a"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 1
    stride: 2
    temporal_stride: 1
  }
}
# ------------- 2nd layer group --------------
layers {
  name: "conv2a"
  type: CONVOLUTION3D
  bottom: "pool1"
  top: "conv2a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 128
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu2a"
  type: RELU
  bottom: "conv2a"
  top: "conv2a"
}
layers {
  name: "pool2"
  type: POOLING3D
  bottom: "conv2a"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}
# ----------------- 3rd layer group --------------
layers {
  name: "conv3a"
  type: CONVOLUTION3D
  bottom: "pool2"
  top: "conv3a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu3a"
  type: RELU
  bottom: "conv3a"
  top: "conv3a"
}
layers {
  name: "pool3"
  type: POOLING3D
  bottom: "conv3a"
  top: "pool3"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}

# --------- 4th layer group
layers {
  name: "conv4a"
  type: CONVOLUTION3D
  bottom: "pool3"
  top: "conv4a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu4a"
  type: RELU
  bottom: "conv4a"
  top: "conv4a"
}
layers {
  name: "pool4"
  type: POOLING3D
  bottom: "conv4a"
  top: "pool4"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}

# --------------- 5th layer group --------
layers {
  name: "conv5a"
  type: CONVOLUTION3D
  bottom: "pool4"
  top: "conv5a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu5a"
  type: RELU
  bottom: "conv5a"
  top: "conv5a"
}
layers {
  name: "pool5"
  type: POOLING3D
  bottom: "conv5a"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}
# ---------------- fc layers -------------
layers {
  name: "fc6"
  type: INNER_PRODUCT
  bottom: "pool5"
  top: "fc6"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 2048
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu6"
  type: RELU
  bottom: "fc6"
  top: "fc6"
}
layers {
  name: "drop6"
  type: DROPOUT
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layers {
  name: "fc7"
  type: INNER_PRODUCT
  bottom: "fc6"
  top: "fc7"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 2048
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu7"
  type: RELU
  bottom: "fc7"
  top: "fc7"
}
layers {
  name: "drop7"
  type: DROPOUT
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layers {
  name: "fc8"
  type: INNER_PRODUCT
  bottom: "fc7"
  top: "fc8"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 101
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layers {
  name: "loss"
  type: SOFTMAX_LOSS
  bottom: "fc8"
  bottom: "label"
}

c. conv3d_ucf101_test.prototxt

name: "deep_c3d_ucf101"
layers {
  name: "data"
  type: VIDEO_DATA
  top: "data"
  top: "label"
  image_data_param {
    source: "../c3d_finetuning/test_01.lst"
    use_image: true
    mean_file: "ucf101_train_mean.binaryproto"
    batch_size: 30
    crop_size: 112
    mirror: false
    show_data: 0
    new_height: 128
    new_width: 171
    new_length: 16
    shuffle: true
  }
}
# ----------- 1st layer group ---------------
layers {
  name: "conv1a"
  type: CONVOLUTION3D
  bottom: "data"
  top: "conv1a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 64
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layers {
  name: "relu1a"
  type: RELU
  bottom: "conv1a"
  top: "conv1a"
}
layers {
  name: "pool1"
  type: POOLING3D
  bottom: "conv1a"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 1
    stride: 2
    temporal_stride: 1
  }
}
# ------------- 2nd layer group --------------
layers {
  name: "conv2a"
  type: CONVOLUTION3D
  bottom: "pool1"
  top: "conv2a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 128
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu2a"
  type: RELU
  bottom: "conv2a"
  top: "conv2a"
}
layers {
  name: "pool2"
  type: POOLING3D
  bottom: "conv2a"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}
# ----------------- 3rd layer group --------------
layers {
  name: "conv3a"
  type: CONVOLUTION3D
  bottom: "pool2"
  top: "conv3a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu3a"
  type: RELU
  bottom: "conv3a"
  top: "conv3a"
}
layers {
  name: "pool3"
  type: POOLING3D
  bottom: "conv3a"
  top: "pool3"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}

# --------- 4th layer group
layers {
  name: "conv4a"
  type: CONVOLUTION3D
  bottom: "pool3"
  top: "conv4a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu4a"
  type: RELU
  bottom: "conv4a"
  top: "conv4a"
}
layers {
  name: "pool4"
  type: POOLING3D
  bottom: "conv4a"
  top: "pool4"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}

# --------------- 5th layer group --------
layers {
  name: "conv5a"
  type: CONVOLUTION3D
  bottom: "pool4"
  top: "conv5a"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  convolution_param {
    num_output: 256
    kernel_size: 3
    kernel_depth: 3
    pad: 1
    temporal_pad: 1
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu5a"
  type: RELU
  bottom: "conv5a"
  top: "conv5a"
}
layers {
  name: "pool5"
  type: POOLING3D
  bottom: "conv5a"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 2
    kernel_depth: 2
    stride: 2
    temporal_stride: 2
  }
}
# ---------------- fc layers -------------
layers {
  name: "fc6"
  type: INNER_PRODUCT
  bottom: "pool5"
  top: "fc6"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 2048
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu6"
  type: RELU
  bottom: "fc6"
  top: "fc6"
}
layers {
  name: "drop6"
  type: DROPOUT
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layers {
  name: "fc7"
  type: INNER_PRODUCT
  bottom: "fc6"
  top: "fc7"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 2048
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layers {
  name: "relu7"
  type: RELU
  bottom: "fc7"
  top: "fc7"
}
layers {
  name: "drop7"
  type: DROPOUT
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layers {
  name: "fc8"
  type: INNER_PRODUCT
  bottom: "fc7"
  top: "fc8"
  blobs_lr: 1
  blobs_lr: 2
  weight_decay: 1
  weight_decay: 0
  inner_product_param {
    num_output: 101
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layers {
  name: "prob"
  type: SOFTMAX
  bottom: "fc8"
  top: "prob"
}
layers {
  top: "accuracy"
  name: "accuracy"
  type: ACCURACY
  bottom: "prob"
  bottom: "label"
}