From 76fa0b09e0d1d638336bafd20750118db66ea9c8 Mon Sep 17 00:00:00 2001 From: Kaushik B <45285388+kaushikb11@users.noreply.github.com> Date: Tue, 16 Feb 2021 13:56:36 +0530 Subject: [PATCH] Update docs/source/reference/object_detection.rst Co-authored-by: Jirka Borovec --- docs/source/reference/object_detection.rst | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/docs/source/reference/object_detection.rst b/docs/source/reference/object_detection.rst index 99a506a167..2840923ca0 100644 --- a/docs/source/reference/object_detection.rst +++ b/docs/source/reference/object_detection.rst @@ -107,7 +107,12 @@ To tailor the object detector to your dataset, you would need to have it in `COC Model ***** -By default, we use the `Faster R-CNN `_ model with a ResNet-50 FPN backbone. We also support `RetinaNet `_. The inputs could be images of different sizes. The model behaves differently for training and evaluation. For training, it expects both the input tensors as well as the targets. And during evaluation, it expects only the input tensors and returns predictions for each image. The predictions are a list of boxes, labels and scores. +By default, we use the `Faster R-CNN `_ model with a ResNet-50 FPN backbone. +We also support `RetinaNet `_. +The inputs could be images of different sizes. +The model behaves differently for training and evaluation. +For training, it expects both the input tensors as well as the targets. And during the evaluation, it expects only the input tensors and returns predictions for each image. +The predictions are a list of boxes, labels, and scores. ------