# Polyp segmentation and neoplasm detection on NeoPolyp-Clean (test)

Source: [Polyp segmentation and neoplasm detection on NeoPolyp-Clean (test)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test>)

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

Data updated: 2026-02-26T05:40:58.776Z

Evaluation of polyp segmentation and neoplasm detection performance on the test split of the NeoPolyp-Clean dataset.

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 | Polyp segmentation and neoplasm detection |
| Dataset (including any stated split) | NeoPolyp-Clean (test) |
| Metric | Dice (Segmentation) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 36 of 36 |
| 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

- [Dice (Segmentation)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=Dice%20(Segmentation)>): Higher is better; unit: SCALAR.

- [IoU (Segmentation)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=IoU%20(Segmentation)>): Higher is better; unit: SCALAR.

- [Dice (Non-neoplastic)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=Dice%20(Non-neoplastic)>): Higher is better; unit: SCALAR.

- [IoU (Non-neoplastic)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=IoU%20(Non-neoplastic)>): Higher is better; unit: SCALAR.

- [Dice (Neoplasm)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=Dice%20(Neoplasm)>): Higher is better; unit: SCALAR.

- [IoU (Neoplasm)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=IoU%20(Neoplasm)>): Higher is better; unit: SCALAR.

- [FPS (Frames Per Second)](<https://www.sota2.com/research/sota/polyp-segmentation-and-neoplasm-detection-on-neopolyp-clean-test.md?metric=FPS%20(Frames%20Per%20Second)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Dice (Segmentation) | Result date | Report paper |
| --- | --- | --- | --- |
| NeoUNet | 0.911 | Not recorded | 68fa6a541e209131cea276c7 |
| NeoUNet | 0.911 | Not recorded | 68fa6bdf1e209131cea30c19 |
| NeoUNet | 0.909 | Not recorded | 68fa6bdf1e209131cea30c19 |
| NeoUNet | 0.908 | Not recorded | 68fa6bdf1e209131cea30c19 |
| BlazeNeo | 0.906 | Not recorded | 68fa6bdf1e209131cea30c19 |
| BlazeNeo | 0.906 | Not recorded | 68fa6bdf1e209131cea30c19 |
| BlazeNeo-DHA | 0.904 | Not recorded | 68fa6bdf1e209131cea30c19 |
| PraNet | 0.895 | Not recorded | 68fa6a541e209131cea276c7 |
| PraNet | 0.895 | Not recorded | 68fa6bdf1e209131cea30c19 |
| HarDNet-MSEG | 0.891 | Not recorded | 68fa6bdf1e209131cea30c19 |
| HarDNet-MSEG | 0.891 | Not recorded | 68fa6bdf1e209131cea30c19 |
| HarDNet-MSEG | 0.883 | Not recorded | 68fa6a541e209131cea276c7 |
| HarDNet-MSEG | 0.883 | Not recorded | 68fa6bdf1e209131cea30c19 |
| BlazeNeo | 0.87 | Not recorded | 68fa6bdf1e209131cea30c19 |
| PraNet | 0.85 | Not recorded | 68fa6bdf1e209131cea30c19 |
| PraNet | 0.849 | Not recorded | 68fa6bdf1e209131cea30c19 |
| NeoUNet | 0.848 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DoubleU-Net | 0.84 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DoubleU-Net | 0.84 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DoubleU-Net | 0.837 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DoubleU-Net | 0.835 | Not recorded | 68fa6bdf1e209131cea30c19 |
| PraNet | 0.817 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DDANet | 0.813 | Not recorded | 68fa6bdf1e209131cea30c19 |
| HarDNet-MSEG | 0.81 | Not recorded | 68fa6bdf1e209131cea30c19 |
| UNet | 0.8 | Not recorded | 68fa6bdf1e209131cea30c19 |
| UNet | 0.8 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DDANet | 0.799 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DDANet | 0.799 | Not recorded | 68fa6bdf1e209131cea30c19 |
| U-Net | 0.785 | Not recorded | 68fa6a541e209131cea276c7 |
| U-Net | 0.785 | Not recorded | 68fa6bdf1e209131cea30c19 |
| DDANet | 0.77 | Not recorded | 68fa6bdf1e209131cea30c19 |
| UNet | 0.764 | Not recorded | 68fa6bdf1e209131cea30c19 |
| ColonSegNet | 0.746 | Not recorded | 68fa6bdf1e209131cea30c19 |
| ColonSegNet | 0.745 | Not recorded | 68fa6bdf1e209131cea30c19 |
| ColonSegNet | 0.738 | Not recorded | 68fa6bdf1e209131cea30c19 |
| ColonSegNet | 0.672 | Not recorded | 68fa6bdf1e209131cea30c19 |

### NeoUNet

Dice (Segmentation): 0.911. Result ID: 698e2a4fffc3901de09b513a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Implementation](<https://github.com/DengPingFan/PraNet>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### NeoUNet

Dice (Segmentation): 0.911. Result ID: 698ea707ffc3901de09f7c05.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Implementation](<https://github.com/DengPingFan/PraNet>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":68.3,"Parameters":"38,288,397","GFLOPS":39.88}. This record does not assert independent reproduction.

### NeoUNet

Dice (Segmentation): 0.909. Result ID: 698ea707ffc3901de09f7bed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Implementation](<https://github.com/DengPingFan/PraNet>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### NeoUNet

Dice (Segmentation): 0.908. Result ID: 698ea707ffc3901de09f7bf5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Implementation](<https://github.com/DengPingFan/PraNet>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### BlazeNeo

Dice (Segmentation): 0.906. Result ID: 698ea707ffc3901de09f7bee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Implementation](<https://github.com/tofuai/neoplasm-segmentation>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### BlazeNeo

Dice (Segmentation): 0.906. Result ID: 698ea707ffc3901de09f7bf6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Implementation](<https://github.com/tofuai/neoplasm-segmentation>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### BlazeNeo-DHA

Dice (Segmentation): 0.904. Result ID: 698ea707ffc3901de09f7c06.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Implementation](<https://github.com/tofuai/neoplasm-segmentation>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":81.5,"Parameters":"17,143,324","GFLOPS":11.06}. This record does not assert independent reproduction.

### PraNet

Dice (Segmentation): 0.895. Result ID: 698e2a4fffc3901de09b5139.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18aea.md>)

[Implementation](<https://github.com/DengPingFan/PraNet1>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PraNet

Dice (Segmentation): 0.895. Result ID: 698ea707ffc3901de09f7c04.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18aea.md>)

[Implementation](<https://github.com/DengPingFan/PraNet1>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":55.6,"Parameters":"30,501,341","GFLOPS":13.11}. This record does not assert independent reproduction.

### HarDNet-MSEG

Dice (Segmentation): 0.891. Result ID: 698ea707ffc3901de09f7beb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69141e209131cea200ce.md>)

[Implementation](<https://github.com/james128333/HarDNet-MSEG>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### HarDNet-MSEG

Dice (Segmentation): 0.891. Result ID: 698ea707ffc3901de09f7bf3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69141e209131cea200ce.md>)

[Implementation](<https://github.com/james128333/HarDNet-MSEG>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### HarDNet-MSEG

Dice (Segmentation): 0.883. Result ID: 698e2a4fffc3901de09b5138.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69141e209131cea200ce.md>)

[Implementation](<https://github.com/james128333/HarDNet-MSEG>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### HarDNet-MSEG

Dice (Segmentation): 0.883. Result ID: 698ea707ffc3901de09f7c03.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69141e209131cea200ce.md>)

[Implementation](<https://github.com/james128333/HarDNet-MSEG>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":77.1,"Parameters":"17,424,031","GFLOPS":11.38}. This record does not assert independent reproduction.

### BlazeNeo

Dice (Segmentation): 0.87. Result ID: 698ea707ffc3901de09f7bfe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Implementation](<https://github.com/tofuai/neoplasm-segmentation>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### PraNet

Dice (Segmentation): 0.85. Result ID: 698ea707ffc3901de09f7bf4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18aea.md>)

[Implementation](<https://github.com/DengPingFan/PraNet1>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### PraNet

Dice (Segmentation): 0.849. Result ID: 698ea707ffc3901de09f7bec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18aea.md>)

[Implementation](<https://github.com/DengPingFan/PraNet1>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### NeoUNet

Dice (Segmentation): 0.848. Result ID: 698ea707ffc3901de09f7bfd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Implementation](<https://github.com/DengPingFan/PraNet>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DoubleU-Net

Dice (Segmentation): 0.84. Result ID: 698ea707ffc3901de09f7bea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d41e209131cea18bd0.md>)

[Implementation](<https://github.com/DebeshJha/2020-CBMS-DoubleU-Net>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DoubleU-Net

Dice (Segmentation): 0.84. Result ID: 698ea707ffc3901de09f7bf2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d41e209131cea18bd0.md>)

[Implementation](<https://github.com/DebeshJha/2020-CBMS-DoubleU-Net>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DoubleU-Net

Dice (Segmentation): 0.837. Result ID: 698ea707ffc3901de09f7c02.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d41e209131cea18bd0.md>)

[Implementation](<https://github.com/DebeshJha/2020-CBMS-DoubleU-Net>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":43.2,"Parameters":"18,836,804","GFLOPS":83.62}. This record does not assert independent reproduction.

### DoubleU-Net

Dice (Segmentation): 0.835. Result ID: 698ea707ffc3901de09f7bfa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d41e209131cea18bd0.md>)

[Implementation](<https://github.com/DebeshJha/2020-CBMS-DoubleU-Net>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### PraNet

Dice (Segmentation): 0.817. Result ID: 698ea707ffc3901de09f7bfc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18aea.md>)

[Implementation](<https://github.com/DengPingFan/PraNet1>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DDANet

Dice (Segmentation): 0.813. Result ID: 698ea707ffc3901de09f7c01.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fa1e209131cea1f764.md>)

[Implementation](<https://github.com/nikhilroxtomar/DDANet>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":46.2,"Parameters":"6,840,000","GFLOPS":31.45}. This record does not assert independent reproduction.

### HarDNet-MSEG

Dice (Segmentation): 0.81. Result ID: 698ea707ffc3901de09f7bfb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69141e209131cea200ce.md>)

[Implementation](<https://github.com/james128333/HarDNet-MSEG>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### UNet

Dice (Segmentation): 0.8. Result ID: 698ea707ffc3901de09f7be8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61e11e209131ce9f653d.md>)

[Implementation](<http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### UNet

Dice (Segmentation): 0.8. Result ID: 698ea707ffc3901de09f7bf0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61e11e209131ce9f653d.md>)

[Implementation](<http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DDANet

Dice (Segmentation): 0.799. Result ID: 698ea707ffc3901de09f7be9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fa1e209131cea1f764.md>)

[Implementation](<https://github.com/nikhilroxtomar/DDANet>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### DDANet

Dice (Segmentation): 0.799. Result ID: 698ea707ffc3901de09f7bf1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fa1e209131cea1f764.md>)

[Implementation](<https://github.com/nikhilroxtomar/DDANet>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### U-Net

Dice (Segmentation): 0.785. Result ID: 698e2a4fffc3901de09b5137.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a541e209131cea276c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61e11e209131ce9f653d.md>)

[Implementation](<http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### U-Net

Dice (Segmentation): 0.785. Result ID: 698ea707ffc3901de09f7c00.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61e11e209131ce9f653d.md>)

[Implementation](<http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":69.6,"Parameters":"31,043,651","GFLOPS":103.59}. This record does not assert independent reproduction.

### DDANet

Dice (Segmentation): 0.77. Result ID: 698ea707ffc3901de09f7bf9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fa1e209131cea1f764.md>)

[Implementation](<https://github.com/nikhilroxtomar/DDANet>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### UNet

Dice (Segmentation): 0.764. Result ID: 698ea707ffc3901de09f7bf8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61e11e209131ce9f653d.md>)

[Implementation](<http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### ColonSegNet

Dice (Segmentation): 0.746. Result ID: 698ea707ffc3901de09f7bef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea2373c.md>)

Recorded evaluation configuration: {"Precision":"FP16","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### ColonSegNet

Dice (Segmentation): 0.745. Result ID: 698ea707ffc3901de09f7be7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea2373c.md>)

Recorded evaluation configuration: {"Precision":"FP32","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

### ColonSegNet

Dice (Segmentation): 0.738. Result ID: 698ea707ffc3901de09f7bff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea2373c.md>)

Recorded evaluation configuration: {"Hardware":"NVIDIA Jetson AGX Xavier","FPS":44.9,"Parameters":"5,010,000","GFLOPS":64.84}. This record does not assert independent reproduction.

### ColonSegNet

Dice (Segmentation): 0.672. Result ID: 698ea707ffc3901de09f7bf7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bdf1e209131cea30c19.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea2373c.md>)

Recorded evaluation configuration: {"Precision":"INT8","Hardware":"NVIDIA Jetson AGX Xavier"}. This record does not assert independent reproduction.

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