API tensorflow detection Faster RCNN with FPN

I would like to use faster RCNN resnet 50 with FPN (feature pyramidal network) so as to get better accuracy on small object. (with ssd, result are worse than faster rcnn). That is why I would try faster rcnn with FPN. But I am a little stuck I do not understand how to add it in the config file. I am currently using the API detection of tensorflow : https://github.com/tensorflow/models/tree/master/research/object_detection

Do I have to change the first stage anchor generator by the multi scale anchor generator ?

Here is the configuration for the FPN if I understand well the protos :

message MultiscaleAnchorGenerator {
  // minimum level in feature pyramid
  optional int32 min_level = 1 [default = 3];

  // maximum level in feature pyramid
  optional int32 max_level = 2 [default = 7];

  // Scale of anchor to feature stride
  optional float anchor_scale = 3 [default = 4.0];

  // Aspect ratios for anchors at each grid point.
  repeated float aspect_ratios = 4;

  // Number of intermediate scale each scale octave
  optional int32 scales_per_octave = 5 [default = 2];

  // Whether to produce anchors in normalized coordinates.
  optional bool normalize_coordinates = 6 [default = true];
}

config file of faster RCNN :

model {
  faster_rcnn {
    num_classes: 4
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 1920
        max_dimension: 1920
      }
    }
    feature_extractor {
      type: "faster_rcnn_resnet50"
      first_stage_features_stride: 16
    }
    first_stage_anchor_generator {
      grid_anchor_generator {
        height_stride: 16
        width_stride: 16
        scales: 0.10000000149011612
        scales: 0.25
        scales: 0.5
        scales: 1.0
        scales: 2.0
        aspect_ratios: 0.25
        aspect_ratios: 0.5
        aspect_ratios: 1.0
        aspect_ratios: 2.0
      }
    }
    first_stage_box_predictor_conv_hyperparams {
      op: CONV
      regularizer {
        l2_regularizer {
          weight: 0.0
        }
      }
      initializer {
        truncated_normal_initializer {
          stddev: 0.009999999776482582
        }
      }
    }
    first_stage_nms_score_threshold: 0.0
    first_stage_nms_iou_threshold: 0.699999988079071
    first_stage_max_proposals: 300
    first_stage_localization_loss_weight: 2.0
    first_stage_objectness_loss_weight: 1.0
    initial_crop_size: 14
    maxpool_kernel_size: 2
    maxpool_stride: 2
    second_stage_box_predictor {
      mask_rcnn_box_predictor {
        fc_hyperparams {
          op: FC
          regularizer {
            l2_regularizer {
              weight: 0.0
            }
          }
          initializer {
            variance_scaling_initializer {
              factor: 1.0
              uniform: true
              mode: FAN_AVG
            }
          }
        }
        use_dropout: false
        dropout_keep_probability: 1.0
      }
    }
    second_stage_post_processing {
      batch_non_max_suppression {
        score_threshold: 0.0
        iou_threshold: 0.6000000238418579
        max_detections_per_class: 100
        max_total_detections: 300
      }
      score_converter: SOFTMAX
    }
    second_stage_localization_loss_weight: 2.0
    second_stage_classification_loss_weight: 1.0
  }
}