Device and a method for associating object detections between frames using a neural network

Patent metadata

Field	Value
Publication number	US-12131518-B2
Application number	US-202117539261-A
Country	US
Kind code	B2
Filing date	Dec 1, 2021
Priority date	Dec 21, 2020
Publication date	Oct 29, 2024
Grant date	Oct 29, 2024

Abstract

A method and a device associate an object detection in a first frame with an object detection in a second frame using a convolutional neural (CNN) network trained to determine feature vectors such that object detections relating to separate objects are arranged in separate clusters. The CNN determines a reference set of feature vectors associated with the object detection in the first frame, and candidate sets of feature vectors associated with a respective one of identified areas corresponding to object detections in the second frame. A set of closest feature vectors is determined, and then measure of closeness to the reference set of feature vectors is determined for each candidate. A respective weight is determined for each object detection in the second frame. The object detection in the first frame is associated with one of the object detections in the second frame based on the assigned weights.

First claim

The invention claimed is:

1. A method for associating an object detection in a first frame with an object detection in a second frame using a convolutional neural network trained to determine feature vectors such that feature vectors of object detections relating to separate objects are arranged in separate clusters in a feature vector space, comprising:
receiving an area of the first frame corresponding to the object detection in the first frame;
determining, by the trained convolutional neural network, a set of feature vectors associated with the area in the first frame comprising two or more feature vectors, thereby determining a reference set of feature vectors associated with the object detection in the first frame comprising two or more feature vectors;
receiving a plurality of areas of the second frame corresponding to respective ones of a plurality of object detections in the second frame;
for each of the plurality of areas in the second frame, determining, by the trained convolutional neural network, a set of feature vectors associated with that area in the second frame comprising two or more feature vectors, thereby determining a plurality of candidate sets of feature vectors, each candidate set of feature vectors comprising two or more feature vectors;
for each feature vector of the reference set of feature vectors, identifying, among the feature vectors of the plurality of candidate sets of feature vectors, a predetermined number of feature vectors having smallest distances in the feature vector space to that feature vector of the reference set of feature vectors, thereby identifying a set of closest feature vectors;
for each candidate set of feature vectors of the plurality of candidate sets of feature vectors, determining a measure of closeness to the reference set of feature vectors as a proportion of the set of closest feature vectors that belong to that candidate set of feature vectors, wherein the predetermined number is set such that the influence of outliers of the reference set and the candidate sets on the measure of closeness is reduced, wherein the outliers of the reference set and the candidate sets are feature vectors that are located at above a threshold distance in the feature vector space away from the other feature vectors of the same of the reference set and the candidate sets, respectively;
for each of the plurality of object detections in the second frame, assigning a weight based on the determined measure of closeness to the reference set of feature vectors of the candidate set of feature vectors corresponding to that object detection; and
associating the object detection in the first frame with one object detection of the plurality of object detections in the second frame based on the assigned weights.

2. The method of claim 1 , wherein:
for any two object detections of the plurality of object detections in the second frame having the same measure of closeness, the assigned weight is the same for the two object detections; and
for any object detection of the plurality of object detections in the second frame having a higher measure of closeness than another object detection of the plurality of object detections in the second frame, the assigned weight is higher for said any object detection than the assigned weight for said another object detection.

3. The method of claim 1 , wherein
for each of the plurality of object detections in the second frame, the assigned weight is proportional to the determined measure of closeness to the reference set of feature vectors of the candidate set of feature vectors corresponding to that object detection.

4. The method of claim 1 , wherein
for each of the plurality object detections in the second frame, the assigned weight is one plus a difference between the measure of closeness of that object detection and the average measure of closeness of the object detections of the plurality of object detections in the second frame other that object detection.

5. The method of claim 1 , further comprising:
for each object detection of the plurality of object detections in the second frame, calculating, based on the weight for that object detection, an association score indicating a probability that that object detection in the second frame is associated with the object detection in the first frame.

6. The method of claim 5 , wherein the association score is further based on a predicted state in the second frame of the object detection in the first frame.

7. The method of claim 6 , wherein the predicted state comprises one or more of a predicted size, predicted shape, predicted location, predicted speed, and acceleration.

8. The method of claim 1 , wherein the object detection in the first frame is associated with an object detection of the plurality of object detections in the second frame having the highest association score.

9. The method of claim 1 , wherein the object detection in first frame is associated to a track, further comprising:
updating the track based on the object detection in second frame with which the object detection in the first frame is associated.

10. The method of claim 1 , wherein the act of determining, by the trained convolutional neural network, a set of feature vectors associated with the area in the first frame comprises:
determining, by the trained convolutional neural network, a first plurality of feature vectors for a plurality of sub-areas of the first image frame; and
identifying a set of feature vectors of the first plurality of feature vectors associated with the area in the first frame comprising two or more feature vectors, thereby determining a reference set of feature vectors comprising two or more feature vectors,
and wherein the act of, for each of the plurality of areas in the second frame, determining, by the trained convolutional neural network, a set of feature vectors associated with that area in the second frame comprises:
determining, by the trained convolutional neural network, a second plurality of feature vectors for a plurality of sub-areas of the second image frame; and
for each of the plurality of areas in the second frame, identifying a set of feature vectors associated with that area in the second frame, thereby determining a plurality of candidate sets of feature vectors.

11. A non-transitory computer-readable storage medium having stored thereon instructions for implementing a method for associating an object detection in a first frame with an object detection in a second frame using a convolutional neural network trained to determine feature vectors such that feature vectors of object detections relating to separate objects are arranged in separate clusters in a feature vector space, when executed by a device having processing capabilities, the method comprising:
receiving an area of the first frame corresponding to the object detection in the first frame;
determining, by the trained convolutional neural network, a set of feature vectors associated with the area in the first frame comprising two or more feature vectors, thereby determining a reference set of feature vectors associated with the object detection in the first frame comprising two or more feature vectors;
receiving a plurality of areas of the second frame corresponding to respective ones of a plurality of object detections in the second frame;
for each of the plurality of areas in the second frame, determining, by the trained convolutional neural network, a set of feature vectors associated with that area in the second frame comprising two or more feature vectors, thereby determining a plurality of candidate sets of feature vectors, each candidate set of feature vectors comprising two or more fe

Assignees

• Axis Ab

Inventors

• Danielsson Niclas
• Liu Haochen

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title

• G06V10/44

  Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components · CPC title

• G06V10/82

  using neural networks · CPC title