# Drivable Region Segmentation on Thermal-Road (test)

Source: [Drivable Region Segmentation on Thermal-Road (test)](<https://www.sota2.com/research/sota/drivable-region-segmentation-on-thermal-road-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:42:22.841Z

Evaluation of Drivable Region Segmentation performance on the test split of the Thermal-Road dataset, measured using segmented Error Rates (campus, mountain, city).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Drivable Region Segmentation |
| Dataset (including any stated split) | Thermal-Road (test) |
| Metric | Error Rate (campus) |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Error Rate (campus)](<https://www.sota2.com/research/sota/drivable-region-segmentation-on-thermal-road-test.md?metric=Error%20Rate%20(campus)>): Lower is better; unit: PERCENT.

- [Error Rate (mountain)](<https://www.sota2.com/research/sota/drivable-region-segmentation-on-thermal-road-test.md?metric=Error%20Rate%20(mountain)>): Lower is better; unit: PERCENT.

- [Error Rate (city)](<https://www.sota2.com/research/sota/drivable-region-segmentation-on-thermal-road-test.md?metric=Error%20Rate%20(city)>): Lower is better; unit: PERCENT.

- [Error Rate (Average)](<https://www.sota2.com/research/sota/drivable-region-segmentation-on-thermal-road-test.md?metric=Error%20Rate%20(Average)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Error Rate (campus) | Result date | Report paper |
| --- | --- | --- | --- |
| PLMp | 9.89 | Not recorded | 68fa639d1e209131ce9ffaeb |
| FCN | 10.3 | Not recorded | 68fa639d1e209131ce9ffaeb |
| TD | 10.7 | Not recorded | 68fa639d1e209131ce9ffaeb |
| PLM | 11.3 | Not recorded | 68fa639d1e209131ce9ffaeb |
| CN24 | 36.4 | Not recorded | 68fa639d1e209131ce9ffaeb |
| AlexNet | 42.1 | Not recorded | 68fa639d1e209131ce9ffaeb |

### PLMp

Error Rate (campus): 9.89. Result ID: 698bfd29f729bc7f68e75bee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

Recorded evaluation configuration: {"Backbone":"Pre-trained with color images","Fine-tuning":"Single initial frame"}. This record does not assert independent reproduction.

### FCN

Error Rate (campus): 10.3. Result ID: 698bfd29f729bc7f68e75bec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa627b1e209131ce9f9766.md>)

[Implementation](<http://fcn.berkeleyvision.org/>)

Recorded evaluation configuration: {"Architecture":"CNN","Supervision":"Fully supervised with Thermal-Road datasets"}. This record does not assert independent reproduction.

### TD

Error Rate (campus): 10.7. Result ID: 698bfd29f729bc7f68e75be9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

Recorded evaluation configuration: {"Feature Type":"Hand-crafted feature based","Description":"Thermal-infrared based drivable region detection"}. This record does not assert independent reproduction.

### PLM

Error Rate (campus): 11.3. Result ID: 698bfd29f729bc7f68e75bed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

Recorded evaluation configuration: {"Backbone":"Pre-trained with color images","Fine-tuning":"Single initial frame"}. This record does not assert independent reproduction.

### CN24

Error Rate (campus): 36.4. Result ID: 698bfd29f729bc7f68e75beb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61b71e209131ce9f5a47.md>)

Recorded evaluation configuration: {"Architecture":"CNN"}. This record does not assert independent reproduction.

### AlexNet

Error Rate (campus): 42.1. Result ID: 698bfd29f729bc7f68e75bea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa639d1e209131ce9ffaeb.md>)

Recorded evaluation configuration: {"Architecture":"CNN"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
