# US 2021/0174133 A1 · Method and device for classifying objects

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0174133 A1
- Document ID: US20210174133A1
- Status: Published application
- Kind: A1
- Application number: 17/181,096
- Filed: 2021-02-22
- Published: 2021-06-10
- Earliest claimed: 2018-09-07
- Assignees: Ibeo Automotive Systems GmbH
- Applicants: Ibeo Automotive Systems GmbH
- Inventors: Julia NITSCH; Max SCHMIDT
- Web page: https://patentgrep.com/patent/US20210174133A1

## CPC

- G06V 20/58: Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06F 18/2413: based on distances to training or reference patterns
- G06F 18/253: {of extracted features}
- G06V 10/454: {Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]}
- G06V 10/806: {of extracted features}
- G06V 10/82: using neural networks
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method for classifying objects which comprises a provision of measuring data from a sensor for a feature extraction unit as well as extraction of modality-independent features from the measuring data by means of the feature extraction unit, wherein the modality-independent features are independent of a sensor modality of the sensor, so that a conclusion to the sensor modality of the sensor is not possible from the modality-independent features.

## Claims

### Claim 1

A method for classifying objects,
wherein the method comprises a provision of measuring data from a sensor for a feature extraction unit,
wherein the method comprises extraction of modality-independent features from the measuring data,
wherein the modality-independent features are independent of a sensor modality of the sensor, so that a conclusion to the sensor modality of the sensor is not possible from the modality-independent features.

### Claim 2 (depends on 1)

The method according to claim 1,
wherein there is at least a first sensor modality and a second sensor modality,
wherein the method is configured to extract modality-independent features from the measuring data from a sensor of the first sensor modality in such a way that the measuring data from a sensor of the second measuring modality can be reconstructed.

### Claim 3 (depends on 1)

The method according to claim 1,
wherein the feature extraction unit comprises at least one feature extractor for each sensor modality,
wherein the feature extractor extracts features from the measuring data.

### Claim 4 (depends on 3)

The method according to claim 3,
wherein the feature extractor is a neural network.

### Claim 5 (depends on 4)

The method according to claim 4,
wherein the first sensor modality is lidar,
wherein the feature extractor of the first sensor modality is a multi layer perceptron (MLP) network.

### Claim 6 (depends on 3)

The method according to claim 3,
wherein the second sensor modality is an image,
wherein the feature extractor of the second sensor modality is a convolutional neural network.

### Claim 7 (depends on 3)

The method according to claim 3,
wherein the feature extraction unit comprises a feature transformation unit,
wherein the feature transformation unit comprises a neural network for each sensor modality, and
wherein the feature transformation unit extracts modality-independent features from features of at least two feature extractors.

### Claim 8 (depends on 1)

The method according to claim 1,
wherein the method comprises the previous training of the feature extraction unit by means of unmonitored learning.

### Claim 9 (depends on 1)

The method according to claim 1,
wherein the method comprises classification of the modality-independent features by means of a classification unit.

### Claim 10 (depends on 9)

The method according to claim 9,
wherein the method comprises the previous training of the classification unit by means of monitored learning.

### Claim 11 (depends on 9)

The method according to claim 9,
wherein the method comprises transferring of at least one feature vector from the feature extraction unit to the classification unit,
wherein the classification comprises comparison of a received feature vector to a respective previously determined average feature vector for each class, and
wherein a corresponding class label is issued when falling below a previously specified deviation limit.

### Claim 12

A method for the optical distance measurement,
wherein the method for the optical distance measurement comprises a method for classifying objects,
wherein said method for classifying objects comprises a provision of measuring data from a sensor for a feature extraction unit and extraction of modality-independent features from the measuring data,
wherein the modality-independent features are independent of a sensor modality of the sensor, so that a conclusion to the sensor modality of the sensor is not possible from the modality-independent features.

### Claim 13

A device for classifying objects,
wherein the device is configured to carry out a method for classifying objects wherein said method for classifying objects comprises provision of measuring data from a sensor for a feature extraction unit and extraction of modality-independent features from the measuring data,
wherein the modality-independent features are independent of a sensor modality of the sensor, so that a conclusion to the sensor modality of the sensor is not possible from the modality-independent features.

### Claim 14

A non-transitory computer-readable storage medium, on which a program is stored, which, after it was loaded into memory of a computer, makes it possible for
the computer to carry out a method
for classifying objects wherein said method for classifying objects comprises provision of measuring data from a sensor for a feature extraction unit and extraction of modality-independent features from the measuring data,
wherein the modality-independent features are independent of a sensor modality of the sensor, so that a conclusion to the sensor modality of the sensor is not possible from the modality-independent features.
