> For the complete documentation index, see [llms.txt](https://yall.yassrobotics.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://yall.yassrobotics.com/documentation/understanding/target-and-pipeline-types.md).

# Target & Pipeline Types

A Limelight pipeline produces exactly one kind of target result, and `LimelightResults` exposes each kind as its own typed array. Only the array matching the active pipeline's type is populated for a given frame; the rest are empty.

| Pipeline type       | `LimelightResults` field | Type                    | Produced by                                                                |
| ------------------- | ------------------------ | ----------------------- | -------------------------------------------------------------------------- |
| AprilTag / fiducial | `targets_Fiducials`      | `AprilTagFiducial[]`    | AprilTag detection + SolvePnP.                                             |
| Retroreflective     | `targets_Retro`          | `RetroreflectiveTape[]` | Color/contour-based retroreflective tape pipelines.                        |
| Neural classifier   | `targets_Classifier`     | `NeuralClassifier[]`    | Whole-frame/zone classification model; labels the scene, no bounding box.  |
| Neural detector     | `targets_Detector`       | `NeuralDetector[]`      | Object detection model; per-object bounding boxes with class + confidence. |
| Barcode             | `targets_Barcode`        | `Barcode[]`             | QR/DataMatrix/etc. decoding.                                               |

Check `LimelightResults.pipeline_type` if your code needs to confirm which pipeline produced a given frame before reading its target array.

## Classifier vs. Detector

These are easy to confuse since both are "neural" pipelines:

* **Classifier** (`NeuralClassifier`): answers "what is this?" for the whole frame or a zone. Has `className`, `confidence`, and a rough `tx`/`ty` position, but no bounding box or area.
* **Detector** (`NeuralDetector`): answers "what objects are where?" with individual bounding boxes. Has `className`, `confidence`, `ta` (area), and crosshair-relative `tx`/`ty`, laid out the same way as `AprilTagFiducial`/`RetroreflectiveTape`. The raw-NT equivalent, `RawDetection`, also exposes the four bounding-box corner points.

## AprilTag vs. Retroreflective: shared pose API

`AprilTagFiducial` and `RetroreflectiveTape` expose the identical set of `getXXXPose_YYYSpace()` / `...2D()` methods (camera-in-target-space, robot-in-field-space, robot-in-target-space, target-in-camera-space, target-in-robot-space). The only real difference is that `AprilTagFiducial` additionally carries `fiducialID` and `fiducialFamily`. If you're writing code that only cares about "where is this target," it can be written generically against either type's pose methods.

## Barcode

`Barcode` is the odd one out: it has no pose accessors, since a printed barcode has no known real-world dimension for SolvePnP to use. It gives you `family`, `data` (the decoded string), pixel/degree position, and optional pixel-space `corners` (must be enabled in the Limelight output tab).

## See Also

* [Object Detection tutorial](/documentation/tutorials/object-detection.md)
* [Tuning a Retroreflective Pipeline](/documentation/tuning/retroreflective-pipeline-tuning.md)
* [Tuning a Neural Network Pipeline](/documentation/tuning/neural-network-pipeline-tuning.md)
* LimelightResults reference


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://yall.yassrobotics.com/documentation/understanding/target-and-pipeline-types.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.