Enhanced object detection for autonomous vehicles based on field view
Systems and methods for enhanced object detection for autonomous vehicles based on field of view. An example method includes obtaining an image from an image sensor of one or more image sensors positioned about a vehicle. A field of view for the image is determined, with the field of view being associated with a vanishing line. A crop portion corresponding to the field of view is generated from the image, with a remaining portion of the image being downsampled. Information associated with detected objects depicted in the image is outputted based on a convolutional neural network, with detecting objects being based on performing a forward pass through the convolutional neural network of the crop portion and the remaining portion.
1. A method implemented by a system of one or more processors, the method comprising:
obtaining a set of images from of a plurality of image sensors positioned about a vehicle, wherein image information from a subset of image sensors is associated with a forward direction of the vehicle;
determining a field of view associated with the image information, the field of view being associated with a vanishing line, wherein the field of view is determined, at least in part, based on an attention-based network and the image information;
generating, based on the attention-based network, information comprising a crop portion corresponding to the field of view, and a remaining portion, wherein the remaining portion is downsampled; and
outputting, via a neural network based, at least in part, on the generated information, information associated with detected objects, wherein detecting objects comprises performing a forward pass through the neural network of the generated information.
2. The method of claim 1 , wherein the neural network receives additional data as input, and wherein the additional data includes data from an inertial measurement unit.
3. The method of claim 1 , wherein the neural network receives additional data as input, and wherein the additional data includes map data.
4. The method of claim 1 , wherein the output information associated with detected objects combines detected objects from the crop portion and the remaining portion.
5. The method of claim 1 , wherein output from neural network is associated with a combination of output associated with the set of images.
6. The method of claim 1 , wherein the crop portion is aligned with the remaining portion.
7. The method of claim 1 , wherein the vanishing line is indicative of a horizon line.
8. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the processors to perform operations comprising:
obtaining a set of images from of a plurality of image sensors positioned about a vehicle, wherein image information from a subset of image sensors is associated with a forward direction of the vehicle;
determining a field of view associated with the image information, the field of view being associated with a vanishing line, wherein the field of view is determined, at least in part, based on an attention-based network and the image information;
generating, based on the attention-based network, information comprising a crop portion corresponding to the field of view, and a remaining portion, wherein the remaining portion is downsampled; and
outputting, via a neural network based, at least in part, on the generated information, information associated with detected objects, wherein detecting objects comprises performing a forward pass through the neural network of the generated information.
9. The system of claim 8 , wherein the neural network receives additional data as input, and wherein the additional data includes data from an inertial measurement unit.
10. The system of claim 8 , wherein the neural network receives additional data as input, and wherein the additional data includes map data.
11. The system of claim 8 , wherein the output information associated with detected objects combines detected objects from the crop portion and the remaining portion.
12. The system of claim 8 , wherein output from neural network is associated with a combination of output associated with the set of images.
13. The system of claim 8 , wherein the crop portion is aligned with the remaining portion.
14. The system of claim 8 , wherein the vanishing line is indicative of a horizon line.
15. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the one or more processors to perform operations comprising:
obtaining a set of images from of a plurality of image sensors positioned about a vehicle, wherein image information from a subset of image sensors is associated with a forward direction of the vehicle;
determining a field of view associated with the image information, the field of view being associated with a vanishing line, wherein the field of view is determined, at least in part, based on an attention-based network and the image information;
generating, based on the attention-based network, information comprising a crop portion corresponding to the field of view, and a remaining portion, wherein the remaining portion is downsampled; and
outputting, via a neural network based, at least in part, on the generated information, information associated with detected objects, wherein detecting objects comprises performing a forward pass through the neural network of the generated information.
16. The computer storage media of claim 15 , wherein the neural network receives additional data as input, and wherein the additional data includes data from an inertial measurement unit and/or map data.
17. The computer storage media of claim 15 , wherein the output information associated with detected objects combines detected objects from the crop portion and the remaining portion.
18. The computer storage media of claim 15 , wherein the vanishing line is indicative of a horizon line.
19. The computer storage media of claim 15 , wherein output from neural network is associated with a combination of output associated with the set of images.
20. The computer storage media of claim 15 , wherein the crop portion is aligned with the remaining portion.