Emigdio Malaver

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

3D Sensor

Image

Video

Top Task Types

Bounding Box

Cuboid

Polygon

Polyline

Freelancer Overview

I worked for 4 years at Remotosks, mainly on projects related to self-driven vehicles. I stayed there as a member of the LATAM elite group, serving as an attempter and reviewer at levels L0 and L10. I am very skilled at box and cuboid annotation and good at polygon annotation, but I need to improve my speed a bit. During the last year I have been working exclusively on 3D projects.

Labeling Experience

Remotasks

Scale AI3D SensorCuboid

This project uses cuboids to annotate each vehicle or pedestrian in the scene. In this project, the ego vehicle is a truck with its trailer, so special attention must be paid to vehicles traveling behind and in the ego's lane, as they enter the sensors' dark zone. Roll: attempter.

This project uses cuboids to annotate each vehicle or pedestrian in the scene. In this project, the ego vehicle is a truck with its trailer, so special attention must be paid to vehicles traveling behind and in the ego's lane, as they enter the sensors' dark zone. Roll: attempter.

2025

Remotasks

Scale AI3D SensorPolygonPolyline

In a 150-frame video, construction zones and the elements used to delimit those zones had to be identified and annotated. Additionally, construction vehicles in the scene had to be also annotated. All these annotations were made using polygons. It was also necessary to adjust the height of the points within these polygons using the Z-editor. The end of the valid data was marked using a polyline. Roll: Reviewer L10.

In a 150-frame video, construction zones and the elements used to delimit those zones had to be identified and annotated. Additionally, construction vehicles in the scene had to be also annotated. All these annotations were made using polygons. It was also necessary to adjust the height of the points within these polygons using the Z-editor. The end of the valid data was marked using a polyline. Roll: Reviewer L10.

2025 - 2025

Romotasks

Scale AI3D SensorPolygonPoint Key Point

In a video with 150 frames, the road markings had to be annotated using polylines, and pedestrian crossings and parking areas using polygons. Tunnel entrances and exits were indicated with a keypoint. The height of each point on the polyline or polygon had to be adjusted separately using the Z-editor. Roll: attempter.

In a video with 150 frames, the road markings had to be annotated using polylines, and pedestrian crossings and parking areas using polygons. Tunnel entrances and exits were indicated with a keypoint. The height of each point on the polyline or polygon had to be adjusted separately using the Z-editor. Roll: attempter.

2025 - 2025

Remotask

Scale AIVideoBounding Box

In a video with 150 frames, traffic channeling devices such as cones, poles, and traffic signs had to be annotated using a single ID. Lidarlite was used as the interface for this project; however, the annotations were made on a 2D projection.

In a video with 150 frames, traffic channeling devices such as cones, poles, and traffic signs had to be annotated using a single ID. Lidarlite was used as the interface for this project; however, the annotations were made on a 2D projection.

2024 - 2024

Remotasks

Scale AIVideoBounding Box

Given a video of 150 frames, the task was to identify the vehicles present using a unique ID, assigning the vehicle type as the only attribute. The unique aspect of this project was that the video had been recorded at night under very low-light conditions.

Given a video of 150 frames, the task was to identify the vehicles present using a unique ID, assigning the vehicle type as the only attribute. The unique aspect of this project was that the video had been recorded at night under very low-light conditions.

2024 - 2024

Show more

Education

Work History