ROOM LAYOUT ESTIMATION METHODS AND TECHNIQUES
Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.
1 . A method for estimating a layout of a room comprising:
accessing a room image for room layout estimation;
determining, using an encoder-decoder sub-network of a neural network for estimating a layout of a room and the room image, a plurality of predicted two-dimensional (2D) keypoint maps corresponding to a plurality of room types;
determining, using the encoder-decoder sub-network, a classifier sub-network of the neural network connected to the encoder-decoder sub-network, and the room image, a predicted room type from the plurality of room types;
determining, using the plurality of predicted 2D keypoint maps and the predicted room type, a plurality of ordered keypoints associated with the predicted room type; and
determining, using the plurality of ordered keypoints, a predicted layout of the room in the room image.
2 . The method of claim 1 , wherein each room type in the plurality of room types comprises an ordered set of room type keypoints.
3 . The method of claim 2 , wherein each room type in the plurality of room types comprises a semantic segmentation for regions in the room type, the semantic segmentation comprising an identification as a floor, a ceiling, or a wall.
4 . The method of claim 2 , wherein a first keypoint order is associated with a first room type of the plurality of room types and a second keypoint order is associated with a second room type of the plurality of room types, wherein the first keypoint order and the second keypoint order are different.
5 . The method of claim 1 , wherein the room image comprises a monocular image.
6 . The method of claim 1 , wherein the room image comprises a Red-Green-Blue (RGB) image.
7 . The method of claim 1 , wherein a dimensionality of the room image is larger than a dimensionality of the predicted 2D keypoint maps.
8 . The method of claim 1 , wherein the encoder-decoder sub-network comprises an encoder sub-network comprising a plurality of convolutional layers and a plurality of pooling layers.
9 . The method of claim 1 , wherein the encoder-decoder sub-network comprises a decoder sub-network comprising a plurality of convolutional layers and a plurality of upsampling layers.
10 . The method of claim 1 , wherein the encoder-decoder sub-network comprises a memory augmented recurrent encoder-decoder (MRED) network.
11 . The method of claim 1 , wherein the encoder-decoder sub-network comprises a plurality of recurrent layers.
12 . The method of claim 11 , wherein a number of recurrent iterations of the plurality of recurrent layers is two.
13 . The method of claim 11 , wherein a number of recurrent iterations of the plurality of recurrent layers is at least three.
14 . The method of claim 11 , wherein each of the plurality of recurrent layers has a weight matrix, and the weight matrix is the same for all of the plurality of recurrent layers.
15 . The method of claim 1 , wherein the predicted two-dimensional (2D) keypoint maps comprise heat maps.
16 . The method of claim 15 , further comprising extracting keypoint locations from the heat maps as maxima of the heat map.
17 . The method of claim 1 , further comprising:
accessing object information from an object recognizer that analyzes the room image; and
combining the object information with the predicted layout of the room.
18 . The method of claim 17 , further comprising:
using the object recognizer to detect cuboids in the room image.
19 . A method for training a neural network for estimating a layout of a room comprising:
receiving a training room image, wherein the training room image is associated with a reference room type from a plurality of room types, and reference keypoints associated with a reference room layout;
generating a neural network for room layout estimation, wherein the neural network comprises an encoder-decoder sub-network configured to output predicted two-dimensional (2D) keypoints associated with a predicted room layout associated with each of the plurality of room types, and a side sub-network connected to the encoder-decoder network configured to output a predicted room type from the plurality of room types; and
optimizing a loss function based on a first loss for the predicted 2D keypoints and a second loss for the predicted room type; and
updating parameters of the neural network based on the optimized loss function.
20 . The method of claim 19 , wherein a number of the plurality of room types is greater than 5.
21 . The method of claim 19 , wherein the reference keypoints and the predicted 2D keypoints are associated with a keypoint order.
22 . The method of claim 19 , wherein a first keypoint order is associated with a first room type of the plurality of room types and a second keypoint order is associated with a second room type of the plurality of room types, wherein the first keypoint order and the second keypoint order are different.
23 . The method of claim 19 , wherein the training room image comprises a monocular image.
24 . The method of claim 19 , wherein the training room image comprises a Red-Green-Blue (RGB) image.
25 . The method of claim 19 , wherein a dimensionality of the training room image is larger than a dimensionality of a map associated with the predicted 2D keypoints.
26 . The method of claim 19 , wherein the encoder sub-network and the decoder sub-network comprises a plurality of recurrent layers.
27 . The method of claim 26 , wherein a number of recurrent iterations of the recurrent layers is two or three.
28 . The method of claim 26 , wherein deep supervision is applied to the recurrent layers.
29 . The method of claim 26 , wherein weights associated with a first recurrent iteration of the iterations of the recurrent layers are identical to weights associated with a second recurrent iteration of the current layers.
30 . The method of claim 26 , wherein the plurality of recurrent layers are configured as a memory-augmented recurrent encoder-decoder (MRED) network.
31 . The method of claim 19 , wherein the side sub-network comprises a room type classifier.
32 . The method of claim 19 , wherein the first loss for the predicted 2D keypoints comprises a Euclidean loss between the plurality of reference keypoint locations and the predicted 2D keypoints.
33 . The method of claim 19 , wherein the second loss for the predicted room type comprises a cross entropy loss.
34 . The method of claim 19 , wherein the predicted 2D keypoints are extracted from a predicted heat map.
35 . The method of claim 34 , further comprising:
calculating a reference heat map associated with the reference keypoints of the training image; and
calculating the first loss for the predicted 2D keypoints based on a difference between the predicted heat map and the reference heat map.
36 . The method of claim 35 , wherein the reference heat map comprises a two-dimensional distribution centered at a location for each reference keypoint.
37 . The method of claim 35 , wherein the reference heat map comprises a background away from the reference keypoints and a foreground associated with the reference keypoints, and the hardware processor is programmed to weight gradients in the reference heat map based on a ratio between the foreground and the background.
38 . The method of claim 37 , wherein weighting the gradients in the reference heat map comprises degrading values of pixels in the background by a degrading factor less than one.
39 . A method comprising:
accessing a room image of an environment;
analyzing the room image to determine a predicted layout of a room in the room image, comprising using a neural network to determine an ordered set of two-dimensional (2D) keypoints associated with a room type for the room in the room image; and
providing the room layout based at least partly on the 2D keypoints and the room type.
40 . The method of claim 39 , wherein the neural network comprises a convolutional encoder-decoder network.
41 . The method of claim 40 , wherein the convolutional encoder-decoder network comprises a memory-augmented recurrent encoder-decoder network.
42 . The method of claim 39 , wherein the neural network comprises a classifier configured to determine the room type.
43 . The method of claim 39 , further comprising extracting the ordered set of 2D keypoints from a heat map.
44 . The method of claim 39 , further comprising:
analyzing the room image to determine object information; and
combining the object information with the room layout.