# Image Classification Calibration on FMoW

Source: [Image Classification Calibration on FMoW](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow>)

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

Data updated: 2026-06-01T09:14:48.815Z

Evaluation of model calibration performance (ECE, ACE) for Image Classification on the FMoW 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 | Image Classification Calibration |
| Dataset (including any stated split) | FMoW |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [ECE](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md?metric=ECE>): Lower is better; unit: PERCENT.

- [ECE (w/ TS)](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md?metric=ECE%20(w%2F%20TS)>): Lower is better; unit: PERCENT.

- [CECE](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md?metric=CECE>): Lower is better; unit: PERCENT.

- [ACE](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md?metric=ACE>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FLSD-53 | 53.81 | Not recorded | 68fa7af41e209131cea85f3e |
| DFL | 53.55 | Not recorded | 68fa7af41e209131cea85f3e |
| BSCE-GRA | 53.53 | Not recorded | 68fa7af41e209131cea85f3e |
| MaxEnt M | 53.04 | Not recorded | 68fa7af41e209131cea85f3e |
| CE | 52.31 | Not recorded | 68fa7af41e209131cea85f3e |
| FGR | 51.95 | Not recorded | 68fa7af41e209131cea85f3e |
| Mixup | 51.53 | Not recorded | 68fa7af41e209131cea85f3e |
| LS-0.05 | 50.89 | Not recorded | 68fa7af41e209131cea85f3e |
| AugMix | 27.47 | Not recorded | 68fa7af41e209131cea85f3e |

### FLSD-53

Accuracy: 53.81. Result ID: 6a1d4d17fbc373592c16bd3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a21e209131cea1d496.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### DFL

Accuracy: 53.55. Result ID: 6a1d4d17fbc373592c16bd40.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f681e209131cea45624.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### BSCE-GRA

Accuracy: 53.53. Result ID: 6a1d4d17fbc373592c16bd42.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77bd1e209131cea750a1.md>)

[Implementation](<https://github.com/Jinxu-Lin/BSCE-GRA>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### MaxEnt M

Accuracy: 53.04. Result ID: 6a1d4d17fbc373592c16bd41.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72711e209131cea565b2.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### CE

Accuracy: 52.31. Result ID: 6a1d4d17fbc373592c16bd3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### FGR

Accuracy: 51.95. Result ID: 6a1d4d17fbc373592c16bd43.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### Mixup

Accuracy: 51.53. Result ID: 6a1d4d17fbc373592c16bd3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69951e209131cea22d47.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### LS-0.05

Accuracy: 50.89. Result ID: 6a1d4d17fbc373592c16bd3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67b71e209131cea180d7.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

### AugMix

Accuracy: 27.47. Result ID: 6a1d4d17fbc373592c16bd3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7af41e209131cea85f3e.md>)

Recorded evaluation configuration: {"Backbone":"DenseNet-121"}. This record does not assert independent reproduction.

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