Image Classification
LiteRT
LiteRT
mediapipe
on-device
android
ios
web
python
flutter
vision
efficientnet
efficientnet-lite2
int8
fp32
Instructions to use litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LiteRT
How to use litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2 with LiteRT:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
File size: 10,377 Bytes
12e1bd1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 | ---
license: apache-2.0
library_name: litert
pipeline_tag: image-classification
datasets:
- imagenet-1k
tags:
- mediapipe
- litert
- tflite
- image-classification
- on-device
- android
- ios
- web
- python
- flutter
- vision
- efficientnet
- efficientnet-lite2
- int8
- fp32
- arxiv:1905.11946
---
# MediaPipe EfficientNet-Lite2 (Image Classifier)
EfficientNet-Lite2 1,000-class ImageNet image classification models in full-integer INT8 (`efficientnet_lite2_int8.tflite`, 7.2 MB) and FP32 (`efficientnet_lite2_float32.tflite`, 24.3 MB), offering higher accuracy than EfficientNet-Lite0 for the [MediaPipe Image Classifier (`ImageClassifier`)](https://developers.google.com/edge/mediapipe/solutions/vision/image_classifier) task and standalone [LiteRT](https://developers.google.com/edge/litert) inference.
<div style="border: 1px solid rgba(59, 130, 246, 0.28); background-color: rgba(59, 130, 246, 0.06); border-radius: 8px; padding: 12px 16px; margin: 16px 0;">
<p style="margin: 0 0 6px 0;"><strong>Documentation:</strong> <a href="https://developers.google.com/edge/mediapipe/solutions/vision/image_classifier">MediaPipe Image Classifier Docs</a></p>
<p style="margin: 0 0 6px 0;"><strong>Samples:</strong> <a href="https://github.com/google-ai-edge/mediapipe-samples/tree/main/examples/image_classification/android">Android</a> · <a href="https://github.com/google-ai-edge/mediapipe-samples/tree/main/examples/image_classification/ios">iOS</a> · <a href="https://colab.research.google.com/github/google-ai-edge/mediapipe-samples/blob/main/examples/image_classification/python/image_classifier.ipynb">Python Colab</a> · <a href="https://github.com/google-ai-edge/mediapipe-samples-web/blob/main/src/tasks/image-classifier.ts">Web</a> · <a href="https://github.com/google/flutter-mediapipe/tree/main/samples/vision/image_classifier">Flutter</a></p>
<p style="margin: 0;"><strong>Try live:</strong> <a href="https://google-ai-edge.github.io/mediapipe-samples-web/#/vision/image_classifier">Web Demo</a></p>
</div>
EfficientNet-Lite2 scales up resolution (`[1, 260, 260, 3]` RGB input) and network capacity over EfficientNet-Lite0 while keeping ReLU6 activations and edge-friendly layers.
### Embedded vs. Standalone Label Files
Both `.tflite` files embed `labels_without_background.txt` (**1,000 classes**, `tench` through `toilet tissue`) inside `TFLITE_METADATA`. The companion [`labels.txt`](./labels.txt) contains **1,001 lines** (line 1 is `background`, lines 2–1001 are the 1,000 ImageNet classes).
## Files
| File in Repo Root | Upstream Google Path | Precision | Size | SHA-256 |
|---|---|---|---:|---|
| [`efficientnet_lite2_int8.tflite`](https://huggingface.co/litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2/blob/main/efficientnet_lite2_int8.tflite) | `image_classifier/efficientnet_lite2/int8/latest/efficientnet_lite2.tflite` | `int8` | 7.17 MB (7,165,461 B) | `7fc3a7dbc1cd2833b162a54577afb47ab531d9c4b799ef11d90a4c56fe4ed5ad` |
| [`efficientnet_lite2_float32.tflite`](https://huggingface.co/litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2/blob/main/efficientnet_lite2_float32.tflite) | `image_classifier/efficientnet_lite2/float32/latest/efficientnet_lite2.tflite` | `float32` | 24.33 MB (24,325,765 B) | `94b6c84a5d6de20f932558de213b50e0338dcc06ab83d8d64b12128a476b4b83` |
### Companion Files in Repo Root
| File | Role | Size | SHA-256 |
|---|---|---:|---|
| [`labels.txt`](./labels.txt) | Label map | 10.5 KB (10,484 B) | `536feacc519de3d418de26b2effb4d75694a8c4c0063e36499a46fa8061e2da9` |
### Download
```bash
pip install -U huggingface_hub
hf download litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2 efficientnet_lite2_int8.tflite --local-dir .
hf download litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2 efficientnet_lite2_float32.tflite --local-dir .
```
## Inputs, Outputs, and Model Specifications
### `efficientnet_lite2_int8.tflite`
- Embedded `TFLITE_METADATA`: name `EfficientNet-lite image classifier (quantized)` · version `1` · author `MediaPipe` · packed associated files: `labels_without_background.txt`
| Direction | Tensor | Shape | Type | Notes |
|---|---|---|---|---|
| Input | `images` | `[1, 260, 260, 3]` | `uint8` | image — Input image to be classified. The expected image is 260 x 260, with three channels (red, blue, and green) per pixel. Each element in the tensor is a value between min and max, where (per-channel) min is [0] and max is [255]. — Normalization: `mean=[127.0], std=[128.0]` — Quantization: `scale=0.0123429, zero_point=125` |
| Output | `Softmax` | `[1, 1000]` | `uint8` | probability — Probabilities of the 1000 labels respectively. — Quantization: `scale=0.00390625, zero_point=0` — Labels: `labels_without_background.txt` |
### `efficientnet_lite2_float32.tflite`
- Embedded `TFLITE_METADATA`: name `EfficientNet-lite image classifier` · version `1` · author `MediaPipe` · packed associated files: `labels_without_background.txt`
| Direction | Tensor | Shape | Type | Notes |
|---|---|---|---|---|
| Input | `images` | `[1, 260, 260, 3]` | `float32` | image — Input image to be classified. The expected image is 260 x 260, with three channels (red, blue, and green) per pixel. Each element in the tensor is a value between min and max, where (per-channel) min is [-0.9921875] and max is [1.0]. — Normalization: `mean=[127.0], std=[128.0]` |
| Output | `Softmax` | `[1, 1000]` | `float32` | probability — Probabilities of the 1000 labels respectively. — Labels: `labels_without_background.txt` |
## Use with MediaPipe Tasks
### Python Quickstart
```bash
pip install -U mediapipe huggingface_hub
```
```python
import mediapipe as mp
from huggingface_hub import hf_hub_download
model_path = hf_hub_download(
repo_id="litert-community/MediaPipe-ImageClassifier-EfficientNet-Lite2",
filename="efficientnet_lite2_int8.tflite",
)
BaseOptions = mp.tasks.BaseOptions
ImageClassifier = mp.tasks.vision.ImageClassifier
ImageClassifierOptions = mp.tasks.vision.ImageClassifierOptions
VisionRunningMode = mp.tasks.vision.RunningMode
options = ImageClassifierOptions(
base_options=BaseOptions(model_asset_path=model_path),
running_mode=VisionRunningMode.IMAGE,
max_results=5,
)
with ImageClassifier.create_from_options(options) as runner:
mp_image = mp.Image.create_from_file("input.jpg")
result = runner.classify(mp_image)
for category in result.classifications[0].categories:
print(f"{category.category_name}: {category.score:.3f}")
```
### Cross-Platform Guides & Samples
- **Task overview & options:** [Image Classifier documentation](https://developers.google.com/edge/mediapipe/solutions/vision/image_classifier)
- **Platform setup guides:** [Python](https://developers.google.com/edge/mediapipe/solutions/setup_python) · [Web / JavaScript](https://developers.google.com/edge/mediapipe/solutions/setup_web) · [Android (Kotlin / Java)](https://developers.google.com/edge/mediapipe/solutions/setup_android) · [iOS (Swift)](https://developers.google.com/edge/mediapipe/solutions/setup_ios) · [Flutter (Dart)](https://developers.google.com/edge/mediapipe/solutions/setup_flutter)
- **Runnable sample apps:** [Web Demo](https://google-ai-edge.github.io/mediapipe-samples-web/#/vision/image_classifier) · [Web Source](https://github.com/google-ai-edge/mediapipe-samples-web/blob/main/src/tasks/image-classifier.ts) · [Android](https://github.com/google-ai-edge/mediapipe-samples/tree/main/examples/image_classification/android) · [iOS](https://github.com/google-ai-edge/mediapipe-samples/tree/main/examples/image_classification/ios) · [Python Notebook](https://colab.research.google.com/github/google-ai-edge/mediapipe-samples/blob/main/examples/image_classification/python/image_classifier.ipynb) · [Flutter](https://github.com/google/flutter-mediapipe/tree/main/samples/vision/image_classifier)
## Latency Benchmarks
End-to-end MediaPipe Task pipeline average latency reported by the [Image Classifier documentation](https://developers.google.com/edge/mediapipe/solutions/vision/image_classifier#benchmarks):
| Model Variant | Device | CPU Latency | GPU Latency |
|---|---|---:|---:|
| EfficientNet-Lite2 (int8) | Pixel 6 | 19.43 ms | — |
| EfficientNet-Lite2 (float 32) | Pixel 6 | 44.17 ms | 22.20 ms |
## Intended Use, Limitations, and Fairness
- **Intended applications:** Higher-accuracy on-device classification of prominent everyday subjects across the 1,000 ImageNet-1K synsets where additional latency over EfficientNet-Lite0 is acceptable.
- **Limitations & considerations:**
- Closed-vocabulary ImageNet-1K classifier; assumes a centered dominant subject inside a `260 x 260` crop.
- Not intended for safety-critical, medical, or biometric decisions without domain-specific fine-tuning.
- Google does not publish a separate standalone fairness PDF for EfficientNet-Lite2 on the MediaPipe documentation page.
## Provenance
Every model binary in this repository is byte-identical (unmodified FlatBuffer weights and embedded `TFLITE_METADATA` bytes) to the official artifact served by Google from `storage.googleapis.com/mediapipe-models`.
| Repo File | Upstream URL | GCS Generation | Upstream `Last-Modified` | Pinned Bucket Version |
|---|---|---|---|---|
| `efficientnet_lite2_int8.tflite` | [`efficientnet_lite2.tflite`](https://storage.googleapis.com/mediapipe-models/image_classifier/efficientnet_lite2/int8/latest/efficientnet_lite2.tflite) | `1682480012152147` | Wed, 26 Apr 2023 03:33:32 GMT | `/1/` |
| `efficientnet_lite2_float32.tflite` | [`efficientnet_lite2.tflite`](https://storage.googleapis.com/mediapipe-models/image_classifier/efficientnet_lite2/float32/latest/efficientnet_lite2.tflite) | `1682480010617750` | Wed, 26 Apr 2023 03:33:30 GMT | `/1/` |
## License and Citation
- License: `apache-2.0` — Stated as Apache License, Version 2.0 (`http://www.apache.org/licenses/LICENSE-2.0`) in the embedded `TFLITE_METADATA` block of the `.tflite` model file(s).
### Citation
```bibtex
@inproceedings{tan2019efficientnet,
title = {EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks},
author = {Tan, Mingxing and Le, Quoc V.},
booktitle = {International Conference on Machine Learning (ICML)},
year = {2019},
url = {https://arxiv.org/abs/1905.11946}
}
```
|