| Model Type | Action Recognition |
|---|---|
| Framework | ONNX |
| Algorithm | CNN+LSTM |
| Technical Details | Uses a CNN to extract spatial features from consecutive video frames and an LSTM to model temporal information and classify human actions across the sequence. |
Video Action Recognition
Classifies human actions from video sequences to support activity understanding, workflow analysis, and safety applications.
Leave your message here
Fill the form with the required information
$0.00
Video Action Recognition
app details
Resources
Use Cases
Quality Control
Industrial Safety
Industrial Surveillance
Surveillance
Emotion Recognition
Supported SECO Devices
AI optimized hardware
Fanless embedded PCs
Modules
supported chipsets
Intel
Intel® Core™ i3
Qualcomm
Qualcomm® Dragonwing QCS6490
Qualcomm® Dragonwing QCS5430
NXP
NXP i.MX 8M Plus
NXP i.MX 95
Performance Benchmark
Hardware · Backend
Processing unit
Latency (ms)
Avg. Peak RAM
MediaTek Genio 510 · ONNX
CPU
246.8 ms
359.5 MB
Intel Tiger Lake · OpenVINO
CPU
25.6 ms
1224.9 MB
Intel Tiger Lake · ONNX
CPU
25.2 ms
621.6 MB
NXP i.MX 8M Plus · ONNX
CPU
499.6 ms
341.1 MB
Intel Tiger Lake · OpenVINO
GPU
31.5 ms
2433.1 MB
Qualcomm QCS6490 · ONNX
CPU
228.6 ms
624.1 MB
Human Fall Detection
AI vision application to detect human falls in real time for safety, elderly care, and industrial risk monitoring.
View Model
HRNet for Human Pose Estimation
High-resolution neural network for accurate keypoint detection in human pose estimation tasks.
View Model
Object Detection – YOLOX
Detects and locates multiple object categories in images and video streams using the lightweight YOLOX architecture.
View Model|
|