Abstract: An approach for handling long-tail distribution with data imbalance. Disclosed embodiments improve Repeat Factor Sampling methods by considering both images and bounding boxes (i.e., instances) to generate improved sets of training data. Disclosed embodiments may be useful in image analysis domains, such as object detection and classification.

Abstract: Identifying key frames of a video for use in training a machine learning model is provided. Object detection is performed to identify frames of a video including target classes of objects of interest. Feature extraction is performed on the identified frames to generate raw feature vectors. The feature vectors are compressed into lower dimension vectors. The compressed feature vectors are compressed into a plurality of clusters. The clustered compressed feature vectors are filtered to identify the key frames from each of the plurality of clusters. The key frames may be provided as a representative data set of the video.

Abstract: Systems and methods for training an object-detection machine learning model with teacher and student framework. The training is intended to exploit a large number of unlabeled image or video frames with few labeled image or video frames for semi-supervised video object detection. For example, the object-detection machine learning model can be pre-trained based on labeled video data, and utilizing pre-trained weights, which initializes a teacher model and a student model with the pre-trained weights. The teacher model is trained to generate pseudo-labels for the unlabeled video data. The student model is trained to generated predicted pseudo-labels for the unlabeled video data, wherein the training of the student model is based on (i) the labeled video data and (ii) the pseudo-labels associated with the unlabeled video data.

Abstract: Methods and systems for classifying a long-tail distribution of data. Data deriving from one or more sensors is classified into a plurality of classes by using (i) a feature-extractor backbone model configured to extract features from the data, and (ii) a classifier model configured to classify the data based on the extracted features. The plurality of classes are grouped, with each group assigned to a respective teacher model. Each respective teacher model is trained with the data in its respective group, as well as the feature-extractor backbone model. The outputs of the teacher models are then merged into a final class prediction model configured to classify the data.

Abstract: Identifying key frames of a video for use in training a machine learning model is provided. Object detection is performed to identify frames of a video including target classes of objects of interest. Feature extraction is performed on the identified frames to generate raw feature vectors. The feature vectors are compressed into lower dimension vectors. The compressed feature vectors are compressed into a plurality of clusters. The clustered compressed feature vectors are filtered to identify the key frames from each of the plurality of clusters. The key frames may be provided as a representative data set of the video.

Abstract: A system and method for generating a robust pseudo-label dataset where a labeled source dataset (e.g., video) may be received and used to train a teacher neural network. A pseudo-labeled dataset may then be output from the teacher network and provided to a similarity-aware weighted box fusion (SWBF) algorithm along with an unlabeled dataset. A robust pseudo-label dataset may then be generated by the SWBF algorithm from and used to train a student neural network. The student neural network may also be further tuned using the labeled source dataset. Lastly, the teacher neural network may be replaced using the student neural network. It is contemplated the system and method may be iteratively repeated.

Abstract: Embodiments of the disclosed techniques disclose methods for planning an indoor radio network for a building. In one embodiment, a method comprises preprocessing an image of a floor plan of the building; generating a radio propagation map for the floor plan using the preprocessed image; and determining an indoor radio transmitter distribution for the floor plan using the radio propagation map.

Abstract: A channel estimation method is provided. The method includes the following steps of: receiving an input symbol of an input signal and obtaining several pilot channel gains through calculation; executing an operation of interpolation on the pilot channel gains by a Wiener filter to obtain several data channel gains through calculation; calculating an adaptive alteration for the first and second multi-path statistical characteristic parameters according to the data channel gains and the pilot channel gains, and accordingly having the first and the second multi-path statistical characteristic parameters adjusted; generating an updated Wiener filter according to the adjusted first and second multi-path statistical characteristic parameters to execute an operation of channel estimation on a next input symbol of the input signal.

Abstract: A channel estimation method is provided. The method includes the following steps of: receiving an input symbol of an input signal and obtaining several pilot channel gains through calculation; executing an operation of interpolation on the pilot channel gains by a Wiener filter to obtain several data channel gains through calculation; calculating an adaptive alteration for the first and second multi-path statistical characteristic parameters according to the data channel gains and the pilot channel gains, and accordingly having the first and the second multi-path statistical characteristic parameters adjusted; generating an updated Wiener filter according to the adjusted first and second multi-path statistical characteristic parameters to execute an operation of channel estimation on a next input symbol of the input signal.