ssd.proto 2.05 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
syntax = "proto2";
package object_detection.protos;

import "object_detection/protos/anchor_generator.proto";
import "object_detection/protos/box_coder.proto";
import "object_detection/protos/box_predictor.proto";
import "object_detection/protos/hyperparams.proto";
import "object_detection/protos/image_resizer.proto";
import "object_detection/protos/matcher.proto";
import "object_detection/protos/losses.proto";
import "object_detection/protos/post_processing.proto";
import "object_detection/protos/region_similarity_calculator.proto";

// Configuration for Single Shot Detection (SSD) models.
message Ssd {

  // Number of classes to predict.
  optional int32 num_classes = 1;

  // Image resizer for preprocessing the input image.
  optional ImageResizer image_resizer = 2;

  // Feature extractor config.
  optional SsdFeatureExtractor feature_extractor = 3;

  // Box coder to encode the boxes.
  optional BoxCoder box_coder = 4;

  // Matcher to match groundtruth with anchors.
  optional Matcher matcher = 5;

  // Region similarity calculator to compute similarity of boxes.
  optional RegionSimilarityCalculator similarity_calculator = 6;

  // Box predictor to attach to the features.
  optional BoxPredictor box_predictor = 7;

  // Anchor generator to compute anchors.
  optional AnchorGenerator anchor_generator = 8;

  // Post processing to apply on the predictions.
  optional PostProcessing post_processing = 9;

  // Whether to normalize the loss by number of groundtruth boxes that match to
  // the anchors.
  optional bool normalize_loss_by_num_matches = 10 [default=true];

  // Loss configuration for training.
  optional Loss loss = 11;
}


message SsdFeatureExtractor {
  // Type of ssd feature extractor.
  optional string type = 1;

  // The factor to alter the depth of the channels in the feature extractor.
  optional float depth_multiplier = 2 [default=1.0];

  // Minimum number of the channels in the feature extractor.
  optional int32 min_depth = 3 [default=16];

  // Hyperparameters for the feature extractor.
  optional Hyperparams conv_hyperparams = 4;
}