Abstract: Systems and methods for obtaining attention features are described. Some examples may include: receiving, at a projector of a transformer, a plurality of tokens associated with image features of a first dimensional space; generating, at the projector of the transformer, projected features by concatenating the plurality of tokens with a positional map, the projected features having a second dimensional space that is less than the first dimensional space; receiving, at an encoder of the transformer, the projected features and generating encoded representations of the projected features using self-attention; decoding, at a decoder of the transformer, the encoded representations and obtaining a decoded output; and projecting the decoded output to the first dimensional space and adding the image features of the first dimensional space to obtain attention features associated with the image features.

Abstract: A positioning model optimization method, a positioning method, and a positioning device are provided. The positioning model optimization method includes: inputting a positioning model for a scene, the positioning model including a three-dimensional (3D) point cloud and a plurality of descriptors corresponding to each 3D point in the 3D point cloud; calculating a significance of each 3D point in the 3D point cloud, and if the significance is greater than a predetermined threshold, outputting the 3D point and the plurality of descriptors corresponding to the 3D point to an optimized positioning model for the scene; and outputting the optimized positioning model for the scene.

Abstract: The present disclosure describes techniques for dual-level semantic segmentation. Data may be input to a first segmentation network. The input data comprises an image and label information associated with the image. The image may be captured at nighttime and may comprise a plurality of regions. At least one region among the plurality of regions may be determined based at least in part on output of the first segmentation network. The at least one region of the image may be cropped. The cropped at least one region may be input to a second segmentation network. A final output may be produced based on the output of the first segmentation network and output of the second segmentation network.

Abstract: A positioning model optimization method, an image-based positioning method and positioning device, and a computer-readable storage medium are provided. The positioning model optimization method includes: inputting a positioning model for a scene, the positioning model including a three-dimensional (3D) point cloud and a plurality of descriptors corresponding to each 3D point in the 3D point cloud; determining, for each 3D point in the 3D point cloud, a plurality of neighboring points of the 3D point, and if a distance relationship between each of the plurality of neighboring points and the 3D point is smaller than a predetermined threshold, outputting the 3D point and the plurality of descriptors corresponding to the 3D point to an optimized positioning model for the scene; and outputting the optimized positioning model for the scene.

Abstract: The present disclosure describes techniques of identifying optimal scheme of knowledge distillation (KD) for vision tasks. The techniques comprise configuring a search space by establishing a plurality of pathways between a teacher network and a student network and assigning an importance factor to each of the plurality of pathways; searching the optimal KD scheme by updating the importance factor and parameters of the student network during a process of training the student network; and performing KD from the teacher network to the student network by retraining the student network based at least in part on the optimized importance factors.

Abstract: A super-network comprising a plurality of layers may be generated. Each layer may comprise cells with different structures. A predetermined number of cells from each layer may be selected. A plurality of cells may be generated based on selected cells using a local mutation model, wherein the local mutation model comprises a mutation window for removing redundant edges from each selected cell. Performance of the plurality of cells may be evaluated using a differentiable fitness scoring function. The operations of the generating a plurality of cells using the local mutation model, the evaluating performance of the plurality of cells using the differentiable fitness scoring function and the selecting the subset of cells based on the evaluation results may be iteratively performed until the super-network converges. A search space for each layer may be generated based on a predetermined top number of cells with largest fitness scores after the super-network converges.

Abstract: Systems and methods for obtaining attention features are described. Some examples may include: receiving, at a projector of a transformer, a plurality of tokens associated with image features of a first dimensional space; generating, at the projector of the transformer, projected features by concatenating the plurality of tokens with a positional map, the projected features having a second dimensional space that is less than the first dimensional space; receiving, at an encoder of the transformer, the projected features and generating encoded representations of the projected features using self-attention; decoding, at a decoder of the transformer, the encoded representations and obtaining a decoded output; and projecting the decoded output to the first dimensional space and adding the image features of the first dimensional space to obtain attention features associated with the image features.

Abstract: Systems and methods for searching a search space are disclosed. Some examples may include using a first parallel module including a first plurality of stacked searching blocks and a second plurality of stacked searching blocks to output first feature maps of a first resolution and to output second feature maps of a second resolution. In some examples, a fusion module may include a plurality of searching blocks, where the fusion module is configured to generate multiscale feature maps by fusing one or more feature maps of the first resolution received from the first parallel module with one or more feature maps of the second resolution received from the first parallel module, and wherein the fusion module is configured to output the multiscale feature maps and output third feature maps of a third resolution.