--- language: en license: apache-2.0 library_name: tensorflow tags: - tensorflow - keras - tflite - emotion-recognition - transformer - lstm - mediapipe - computer-vision - deep-learning - facial-expression - affective-computing - sequential-data model-index: - name: emotion_landmark_lstm_model results: - task: type: sequence-classification dataset: type: dataset name: Optimized 478-Point 3D Facial Landmark Dataset metrics: - name: accuracy type: float value: 0.7289 inference: "Supports TensorFlow and TensorFlow Lite real-time inference" --- # πŸŽ₯ Emotion Sequence Transformer (TensorFlow) β€” Mediapipe 478 Landmarks (Seq256) **Version:** v1.0 **Framework:** TensorFlow 2.x **Optimized format:** TensorFlow Lite **Input:** 478 Mediapipe Face Mesh landmarks per frame (up to 300 frames) **Output:** 6-class emotion prediction (`Angry`, `Disgust`, `Fear`, `Happy`, `Neutral`, `Sad`) --- ## 🧠 Model Overview The **Emotion Sequence Transformer** is a deep learning model built using TensorFlow for recognizing **human emotions** from continuous **video clips**. It uses **478 Mediapipe facial landmarks per frame** to capture spatiotemporal patterns of facial movements across time. The model predicts one of six basic emotions by analyzing both facial geometry and temporal variation within sequences of up to **300 frames**. This model is suitable for **real-time video-based emotion detection**, **affective computing**, **human-computer interaction**, and **emotion-aware AI systems**. --- ## πŸ“Š Dataset This model was trained on the **[Optimized 478-Point 3D Facial Landmark Dataset](https://www.kaggle.com/datasets/psewmuthu/optimized-video-facial-landmarks)** β€” a dataset derived from the **Video Emotion Dataset**, optimized for emotion recognition using Mediapipe’s 3D face mesh landmarks. Each sample in the dataset includes: - Up to **300 frames per clip** - **478 facial landmarks per frame** - Corresponding **emotion label** --- ## 🧩 Model Architecture The architecture is based on a **Transformer encoder** design that processes sequential data of facial landmarks. **Pipeline:** 1. Input normalization using precomputed mean and std (global stats) 2. Sequence embedding via positional encodings 3. Transformer encoder blocks to capture temporal and spatial dependencies 4. Dense layers for emotion classification (6 output neurons with softmax) **Core Components:** - Transformer Encoder Layers (Multi-Head Self-Attention) - Layer Normalization and Dropout - Dense classification head --- ## πŸ“ˆ Performance | Metric | Value | | --------------------- | ---------- | | **Test Accuracy** | 0.7289 | | **Test Loss** | 1.1336 | | **Macro F1-Score** | 0.73 | | **Weighted F1-Score** | 0.73 | | **Max Clip Length** | 300 frames | | **Input Shape** | (300, 478) | ### 🧾 Classification Report | Emotion | Precision | Recall | F1-score | Support | | -------------------- | --------- | ------ | ------------------- | ------- | | Angry | 0.75 | 0.73 | 0.74 | 139 | | Disgust | 0.88 | 0.70 | 0.78 | 128 | | Fear | 0.52 | 0.60 | 0.55 | 114 | | Happy | 0.88 | 0.97 | 0.92 | 129 | | Neutral | 0.66 | 0.79 | 0.72 | 101 | | Sad | 0.70 | 0.58 | 0.64 | 134 | | **Overall Accuracy** | **0.73** | | **Macro Avg: 0.73** | 745 | --- ## πŸ“Š Visualizations ### πŸ”Ή Training Accuracy and Loss ![Accuracy and Loss](images/Accuracies_and_Losses.png) ### πŸ”Ή Confusion Matrix ![Confusion Matrix](images/Confusion_Matrix.png) ### πŸ”Ή ROC Curves (Per Class) ![ROC Curves](images/ROC_Curves.png) --- ## πŸ“‚ Repository Structure ``` TF-Emotion-Sequence-Transformer/ β”œβ”€β”€ tf_emotion_sequence_transformer_mp478_seq256.h5 β”œβ”€β”€ tf_emotion_sequence_transformer_mp478_seq256_optimized.tflite β”œβ”€β”€ tf_emotion-sequence-transformer-bilstm-usage.ipynb β”œβ”€β”€ assets/ β”‚ β”œβ”€β”€ global_mean.npy β”‚ β”œβ”€β”€ global_std.npy β”‚ β”œβ”€β”€ label_encoder.pkl β”‚ └── metadata.json └── README.md ``` ### File Descriptions | File | Description | | --------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------- | | `tf_emotion_sequence_transformer_mp478_seq256.h5` | Main TensorFlow model trained on 478 landmarks (300 frames max). | | `tf_emotion_sequence_transformer_mp478_seq256_optimized.tflite` | Optimized TensorFlow Lite version for deployment (mobile, edge). | | `tf_emotion-sequence-transformer-bilstm-usage.ipynb` | Example notebook demonstrating how to use the model for emotion prediction from Mediapipe landmarks. | | `assets/global_mean.npy` | Precomputed global mean for normalization. | | `assets/global_std.npy` | Precomputed global standard deviation for normalization. | | `assets/label_encoder.pkl` | Encoder mapping integer labels to emotion names. | | `assets/metadata.json` | Model metadata and configuration details. | --- ## πŸš€ Example Usage ### πŸ”Έ TensorFlow (.h5) Model ```python import numpy as np import tensorflow as tf import joblib import json # Load Model model = tf.keras.models.load_model("tf_emotion_sequence_transformer_mp478_seq256.h5") # Load assets mean = np.load("assets/global_mean.npy") std = np.load("assets/global_std.npy") label_encoder = joblib.load("assets/label_encoder.pkl") # Preprocess input input_seq = np.load("example_input.npy") # shape: (300, 478) input_seq = (input_seq - mean) / std input_seq = np.expand_dims(input_seq, axis=0) # Predict pred = model.predict(input_seq) emotion = label_encoder.inverse_transform([np.argmax(pred)])[0] print("Predicted Emotion:", emotion) ``` --- ### πŸ”Έ TensorFlow Lite (Optimized) Model ```python import numpy as np import tensorflow as tf import joblib # Load TFLite model interpreter = tf.lite.Interpreter(model_path="tf_emotion_sequence_transformer_mp478_seq256_optimized.tflite") interpreter.allocate_tensors() # Get input and output tensors input_details = interpreter.get_input_details() output_details = interpreter.get_output_details() # Load preprocessing assets mean = np.load("assets/global_mean.npy") std = np.load("assets/global_std.npy") label_encoder = joblib.load("assets/label_encoder.pkl") # Prepare input input_seq = np.load("example_input.npy") # shape: (300, 478) input_seq = (input_seq - mean) / std input_seq = np.expand_dims(input_seq, axis=0).astype(np.float32) # Inference interpreter.set_tensor(input_details[0]['index'], input_seq) interpreter.invoke() pred = interpreter.get_tensor(output_details[0]['index']) # Decode emotion emotion = label_encoder.inverse_transform([np.argmax(pred)])[0] print("Predicted Emotion:", emotion) ``` --- ## πŸ”– Version Information **Version:** v1.0 **Date:** November 2025 **Author:** [P.S. Abewickrama Singhe](https://www.kaggle.com/psewmuthu) **Framework:** TensorFlow 2.x **Exported Models:** `.h5`, `.tflite` **Landmarks per frame:** 478 **Max frames per clip:** 300 --- ## 🏷️ Tags `tensorflow` β€’ `emotion-recognition` β€’ `mediapipe` β€’ `transformer` β€’ `sequence-model` β€’ `facial-landmarks` β€’ `video-analysis` β€’ `tflite` β€’ `human-emotion-ai` β€’ `affective-computing` β€’ `computer-vision` β€’ `deep-learning` --- ## πŸ“š Citation If you use this model in your research, please cite it as: ```bibtex @misc{pasindu_sewmuthu_abewickrama_singhe_2025, author = { Pasindu Sewmuthu Abewickrama Singhe }, title = { EmotionFormer-BiLSTM (Revision f329517) }, year = 2025, url = { https://huggingface.co/PSewmuthu/EmotionFormer-BiLSTM }, doi = { 10.57967/hf/6899 }, publisher = { Hugging Face } } ``` --- ## πŸͺͺ License This model is released under the **Apache 2.0 License** β€” free for academic and commercial use with attribution. ---