Faster R-CNN Models

Construct Faster R-CNN model variants for object detection tasks.

model_fasterrcnn_resnet50_fpn(
  pretrained = FALSE,
  progress = TRUE,
  num_classes = 91,
  ...
)

model_fasterrcnn_resnet50_fpn_v2(
  pretrained = FALSE,
  progress = TRUE,
  num_classes = 91,
  ...
)

model_fasterrcnn_mobilenet_v3_large_fpn(
  pretrained = FALSE,
  progress = TRUE,
  num_classes = 91,
  ...
)

model_fasterrcnn_mobilenet_v3_large_320_fpn(
  pretrained = FALSE,
  progress = TRUE,
  num_classes = 91,
  ...
)

Arguments

pretrained

Logical. If TRUE, loads pretrained weights from local file.

progress

Logical. Show progress bar during download (unused).

num_classes

Number of output classes (default: 91 for COCO).

...

Other arguments (unused).

Value

A fasterrcnn_model nn_module.

Functions

• model_fasterrcnn_resnet50_fpn(): Faster R-CNN with ResNet-50 FPN

• model_fasterrcnn_resnet50_fpn_v2(): Faster R-CNN with ResNet-50 FPN V2

• model_fasterrcnn_mobilenet_v3_large_fpn(): Faster R-CNN with MobileNet V3 Large FPN

• model_fasterrcnn_mobilenet_v3_large_320_fpn(): Faster R-CNN with MobileNet V3 Large 320 FPN

Task

Object detection over images with bounding boxes and class labels.

Object detection over images with bounding boxes and class labels.

Input Format

Input images should be torch_tensors of shape (batch_size, 3, H, W) where H and W are typically around 800.

Available Models

• model_fasterrcnn_resnet50_fpn()

• model_fasterrcnn_resnet50_fpn_v2()

• model_fasterrcnn_mobilenet_v3_large_fpn()

• model_fasterrcnn_mobilenet_v3_large_320_fpn()

See also

Other object_detection_model: model_facenet

Examples

if (FALSE) { # \dontrun{
library(magrittr)
norm_mean <- c(0.485, 0.456, 0.406) # ImageNet normalization constants, see
# https://pytorch.org/vision/stable/models.html
norm_std  <- c(0.229, 0.224, 0.225)
# Use a publicly available image of an animal
wmc <- "https://upload.wikimedia.org/wikipedia/commons/thumb/"
url <- "e/ea/Morsan_Normande_vache.jpg/120px-Morsan_Normande_vache.jpg"
img <- base_loader(paste0(wmc,url))

input <- img %>%
  transform_to_tensor() %>%
  transform_resize(c(520, 520)) %>%
  transform_normalize(norm_mean, norm_std)
batch <- input$unsqueeze(1)    # Add batch dimension (1, 3, H, W)

# ResNet-50 FPN
model <- model_fasterrcnn_resnet50_fpn(pretrained = TRUE)
model$eval()
pred <- model(batch)$detections
num_boxes <- as.integer(pred$boxes$size()[1])
keep <- seq_len(min(5, num_boxes))
boxes <- pred$boxes[keep, ]$view(c(-1, 4))
labels <- ds$category_names[as.character(as.integer(pred$labels[keep]))]
if (num_boxes > 0) {
  boxed <- draw_bounding_boxes(image, boxes, labels = labels)
  tensor_image_browse(boxed)
}

# ResNet-50 FPN V2
model <- model_fasterrcnn_resnet50_fpn_v2(pretrained = TRUE)
model$eval()
pred <- model(batch)$detections
num_boxes <- as.integer(pred$boxes$size()[1])
keep <- seq_len(min(5, num_boxes))
boxes <- pred$boxes[keep, ]$view(c(-1, 4))
labels <- ds$category_names[as.character(as.integer(pred$labels[keep]))]
if (num_boxes > 0) {
  boxed <- draw_bounding_boxes(image, boxes, labels = labels)
  tensor_image_browse(boxed)
}

# MobileNet V3 Large FPN
model <- model_fasterrcnn_mobilenet_v3_large_fpn(pretrained = TRUE)
model$eval()
pred <- model(batch)$detections
num_boxes <- as.integer(pred$boxes$size()[1])
keep <- seq_len(min(5, num_boxes))
boxes <- pred$boxes[keep, ]$view(c(-1, 4))
labels <- ds$category_names[as.character(as.integer(pred$labels[keep]))]
if (num_boxes > 0) {
  boxed <- draw_bounding_boxes(image, boxes, labels = labels)
  tensor_image_browse(boxed)
}

# MobileNet V3 Large 320 FPN
model <- model_fasterrcnn_mobilenet_v3_large_320_fpn(pretrained = TRUE)
model$eval()
pred <- model(batch)$detections
num_boxes <- as.integer(pred$boxes$size()[1])
keep <- seq_len(min(5, num_boxes))
boxes <- pred$boxes[keep, ]$view(c(-1, 4))
labels <- ds$category_names[as.character(as.integer(pred$labels[keep]))]
if (num_boxes > 0) {
  boxed <- draw_bounding_boxes(image, boxes, labels = labels)
  tensor_image_browse(boxed)
}
} # }