Keymakr

Services: Data Labeling (bounding box), pre-annotation validation, quality control

Statistics: 
Over 500,000 data units
Team of 80+ specialists
8x annotation capacity increase with Keylabs.ai


Intro
A multinational enterprise that builds tools for ADAS needed expert annotation of road-scene imagery and video to strengthen AI models. The organization aggregates visual and sensor data from diverse geographies and operating conditions, and requires outputs that could be aligned to its internal pipelines and reviewed under strict quality thresholds.
In this case, the company needed highly accurate traffic imagery and video annotation to train and refine AI models for environmental perception. The goal was to ensure that systems could correctly identify lane markings and conditions across different regions, a critical step in building reliable, real-world-ready mobility solutions.


The challenge
Autonomous vehicle systems depend on precisely labeled training data to interpret and respond to real-world conditions. In this project, the work involved managing multiple layers of image and video data, with variations across many factors.

The complexity was intensified by several key challenges:

Continuous model refinement
The project required regular retraining based on new inputs and field test results. This created an ongoing demand for data validation and labeling, not just any annotation, but expert-level precision aligned with evolving requirements.

Critical precision in an automotive context
In automotive AI, even minor annotation errors can have serious consequences. Mislabeling a traffic sign or failing to notice an animal on the road might mislead a model controlling a self-driving system. Accuracy was vital.

Region-specific data
The annotation team needed to accurately identify and label region-specific data because the system was deployed in multiple regions, each with unique traffic signage, layouts, and movement patterns.

High volume, tight timelines
Workflows often required processing large-scale datasets under tight deadlines — up to 200,000 object instances (appearances of objects in individual images) within a 6-7 day turnaround, all while maintaining strict quality standards.

Scalability bottlenecks
Part of the annotation workflow was initially executed on an open-source tool. However, when scaling was attempted, the system began to lag and limit productivity. A scalable, flexible solution was needed to support large, concurrent teams efficiently.

Results

The project’s annotation workflow was significantly optimized through a structured validation and labeling process built to handle complex automotive datasets.

Scalability limitations were fully resolved through migration to Keylabs.ai. Post-migration, over 50 operators were able to work concurrently without performance degradation, resulting in an 8x increase in operational capacity.

Keylabs’ architecture ensured readiness to scale at any point, enabling the customer to respond to changing data volumes and project phases without workflow disruption.

Labeling speed was improved using Keylabs.ai platform features. Tools such as macroshots enabled annotators to instantly locate and preview specific object classes (e.g., traffic signs), reducing instruction-related errors and increasing throughput.

Throughout the engagement, strict deadlines were consistently met, with a team of over 80 trained specialists working across multiple annotation phases. In total, more than 500,000 data units were processed with high precision, supporting the ongoing development of ADAS and traffic detection systems for real-world deployment.

Industry: Automotive, autonomous driving
Services: 3D labeling (polylines), LiDAR annotation, custom tool development, pipeline customization, data validation

Overview:

Case study period: Q2 2025
Team: 12 specialists
Annotation efficiency: ~21% growth, thanks to custom tools and solutions

Intro

The client is one of the world’s largest tech corporations, ranked among the top 500 companies by market capitalization. It invests heavily in autonomous driving technologies.

The models and algorithms developed by the company enable vehicles to make reliable, real-time decisions. As part of this initiative, the company approached Keymakr to create high-precision 3D labeling of road markings based on LiDAR data, essential for training and validating its autopilot systems.

The challenge
The client required data annotation of road markings and road edges, which later needed to be aligned with geospatial data.

The primary goal was to perform the labeling in 3D, with all work done using polylines. Typically, the Keymakr team uses segmentation or cuboids, but in this case, linear object construction was required.

Key challenges of the project:
 

• Hidden bottleneck of 3D data

Working with spatial polylines in 3D is significantly more complex than working with standard 2D images. LiDAR captures the environment as a point cloud, lacking color and visual cues. Road markings may be partially worn out, obscured by vehicles or pedestrians, blurred, or disappear entirely due to shadows from objects. Under these conditions, an annotator must not simply “draw a line,” but reconstruct the geometry of the road despite incomplete visibility.

One of the key difficulties was occlusion: dynamic and static objects blocked parts of the markings. In a classic 2D workflow, an annotator would simply break the line, outline the occluding object, and then continue the line. However, in 3D, such a break may be interpreted by the autopilot system as the actual end of the line, which can critically affect decision-making while driving.

• Working with aggregated 3D scenes

An additional complexity came from the client’s data format: the scenes were organized not by frames but along a timeline. A static object was technically present at every moment in time, and if processed frame by frame, the file would grow rapidly in size and take up more space. Any manual refinement would have turned into an impractical and slow procedure.

Beyond geometry, it was also necessary to account for semantic relationships between lines: when one line transitioned into another, split, changed type, or merged.

• Measuring confidence level

The project had a unique requirement: the visibility level (confidence level) had to be assigned to every vertex of the polyline. At the same time, the 3D platform only supported attributes at the whole-object level.

• Automatic measurement of occlusions and line breaks

If an occlusion exceeded 10 meters, the line had to be split. Manually measuring such distances was extremely time-consuming: each break had to be measured with a “ruler” tool while switching to the “top view.” This made a custom measurement solution within the converter algorithm essential.

• Reverse сonverter

Once the data was converted into the final format, it needed to be converted back into a format readable by the platform, so that the results could be visually verified. A verification converter needed to be created to finalize the output. 

This project demonstrated how the 3D labeling process evolves into an engineering task at the R&D level, and how working with data leads to the full-scale development of solutions for autonomous driving systems. The outcome became an important milestone not only for the client but also for advancing Keymakr’s approach to 3D annotation of complex road scenes.

Results

Despite the scale of the task and the complexity of the data, the project was delivered on time and with high accuracy.

The most significant outcome was not only the completed labeling itself, but the new technological infrastructure built throughout the project.

These solutions increased labeling speed by approximately 21% and ensured the structural integrity and consistency required for autonomous driving algorithms.

The client received a correct, semantically rich, and geometrically precise 3D annotation dataset fully aligned with their technical KPIs. Meanwhile, Keymakr gained a new suite of tools and methods that can now be applied to other projects involving complex 3D and LiDAR data.