Update README.md

README.md

@@ -2,11 +2,11 @@
 
 [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE)
 
-Media + AI video analytics benchmarking application utilizing [DLStreamer](https://github.com/open-edge-platform/edge-ai-libraries/tree/main/libraries/dl-streamer) and [GStreamer](https://gstreamer.freedesktop.org)*. Measures end-to-end throughput in fps, pipeline stream density, package power, and workload efficiency as FPS per Package Watt.
+Edge Workloads and Benchmarks are performance-optimized pipelines that utilizes the [GStreamer multimedia framework](https://gstreamer.freedesktop.org) and the [Deep Learning Streamer (DL Streamer)](https://github.com/open-edge-platform/edge-ai-libraries/tree/main/libraries/dl-streamer) for validating media and edge AI video analytics. The pipelines measure end-to-end throughput in frames per second (fps), pipeline stream density in fps, package power in watts, and workload efficiency in fps per package watt.
 
 ### Pipeline Architecture
 
-HEVC 1080p Video Decode (GPU HW-Accelerated) → Object Detection (GPU/NPU) → Object Tracking → 1-2x Object Classification (GPU/NPU)
+HEVC 1080p Video Decode (GPU Hardware-Accelerated) → Object Detection (GPU or NPU) → Object Tracking → 1-2x Object Classification (GPU or NPU)
 
 ## Pipeline Configurations
 | Config | Video | Detection | Classification |
@@ -15,42 +15,50 @@ HEVC 1080p Video Decode (GPU HW-Accelerated) → Object Detection (GPU/NPU) →
 | medium | apple.h265 (1 obj/frame) | YOLOv5m (640x640) INT8 | ResNet‑50 + MobileNet‑V2 (224x224) INT8 |
 | heavy | bears.h265 (2 obj/frame) | YOLOv11m (640x640) INT8 | ResNet‑50 + MobileNet‑V2 (224x224) INT8 |
 
-Pipeline configurations include single-device pipelines (GPU/NPU-only), pipelines with multiple devices (GPU Detect + NPU Classify), and multiple single-device pipelines running concurrently in separate processes (GPU-Only + NPU-Only concurrently).
+Pipeline configurations include single-device pipelines (GPU or NPU only), pipelines with multiple devices (GPU Detect and NPU Classify), and multiple single-device pipelines running concurrently in separate processes (GPU only and NPU only, concurrently).
 
 ## Prerequisites
-**System Requirements:**  
-A GPU with VA-API media support is required for this workload.
-- Validated against Ubuntu 24.04.3 LTS with Kernel 6.16+
-- Docker installed and user in docker group
-- Integrated GPU
-- NPU (optional)
-
-
-**Required Software:**
-- Docker 20.10+ ([installation guide](https://docs.docker.com/engine/install/ubuntu/))
-- Python 3.8+ with venv support
-- Network connectivity for model/media download
-
-**Storage Space Requirements:**
-- Models: 230MB
-- COCO dataset: 950MB
-- CIFAR-100: 162MB
-- Videos: 1.9GB
-- Virtual Environment: 7.7GB
-- Total Required: 10.9GB
-
-Optional ImageNet Dataset Download: +6.5GB (manual download required, see [model-conversion](model-conversion/README.md))
-
-**Display Pipeline:**  
-The display pipeline sample requires access to the display. Run the following commands to allow the X server connection in Docker:
+
+### System Requirements
+
+A GPU with video acceleration API (VA-API) media support that:
+- Is validated on Ubuntu OS version 24.04.3 LTS with kernel version 6.16 and above
+- Has Docker software installed and the user added in the docker group
+- Has Integrated GPU
+- Has NPU (optional)
+
+
+### Software Requirements
+
+- Docker software version 20.10 and above ([installation guide](https://docs.docker.com/engine/install/ubuntu/))
+- Python programming language version 3.8 and above with virtual environment (venv) support
+- Network connectivity for model or media download
+
+### Storage Space Requirements
+
+Breakdown of space required:
+- Models: 230 MB
+- COCO dataset: 950 MB
+- CIFAR-100 dataset: 162 MB
+- Videos: 1.9 GB
+- Virtual Environment: 7.7 GB
+
+Total space required: 10.9 GB
+
+Optional ImageNet Dataset Download: 6.5GB. You will need to download this manually. See [model-conversion](model-conversion/README.md).
+
+### Display Pipeline
+
+Run the following commands to allow X server connection in the Docker container, so that the display pipeline sample can access the host's display:
+
 ```bash
 xhost local:root
 setfacl -m user:1000:r ~/.Xauthority
 ```
 
 ## Get Started
 
-The Makefile automates the entire workflow. use `make help` to display the following:
+The Makefile automates the entire workflow. Use `make help` to display the following:
 
 ```bash
 # Three-step setup
@@ -71,49 +79,55 @@ make display          # Visualized pipeline demo (params: CONFIG={light,medium,h
 
 # Cleanup
 make clean            # Remove all results
-make clean-all        # Remove all generated collateral (models, media, results, venv)
+make clean-all        # Remove all generated collateral (models, media, results, and venv)
 ```
 
-#### Examples
+### Examples
+
 - Prereqs: `make prereqs INCLUDE_GPU=True INCLUDE_NPU=True`
 - Display: `make display CONFIG=light DETECT=GPU CLASSIFY=NPU DURATION=60`
 - Benchmarks: `sudo make benchmarks CORES=ecore DURATION=60`
 
-#### Makefile Variables
-- `IMAGENET_ROOT` - Path to pre-downloaded ImageNet dataset for accuracy validation on Resnet and Mobilenet (see [model-conversion](model-conversion/README.md))
+### Makefile Variables
+
+- `IMAGENET_ROOT` - Path to pre-downloaded the ImageNet dataset for accuracy validation on ResNet architecture and MobileNet networks (see [model-conversion](model-conversion/README.md))
 - `INCLUDE_GPU` - Install GPU drivers during setup
 - `INCLUDE_NPU` - Install NPU drivers during setup (requires reboot)
 - `DURATION` - Duration to run pipeline in seconds
 - `CONFIG=light|medium|heavy` - Pipeline configuration, tiered by compute complexity
 - `DETECT/CLASSIFY=CPU|GPU|NPU` - Inference device assignment
-- `CORES=pcore|ecore|lpecore` - CPU core pinning based off of core type
+- `CORES=pcore|ecore|lpecore` - CPU core pinning based on core type
 - `PORT` - HTTP server port for dashboard (default: 8000)
 
 ### Manual Setup (Alternative)
 
 If you prefer step-by-step control:
 
-#### Step 1: Prerequisites
+Step 1. Prerequisites:
+
 ```bash
 cd setup/
 ./install_prerequisites.sh
 # Optional: --reinstall-gpu-driver=yes and/or --reinstall-npu-driver=yes
 ```
 
-#### Step 2: Models
+Step 2. Models:
+
 ```bash
 cd ../model-conversion/
 ./convert_models.sh
 # Optional: -i "$HOME/datasets/imagenet-packages" for ImageNet quantization
 ```
 
-#### Step 3: Media
+Step 3. Media:
+
 ```bash
 cd ../media-downloader/
 ./download_and_encode.sh
 ```
 
-#### Step 4: Run Benchmark
+Step 4. Run benchmark:
+
 ```bash
 cd ..
 ./benchmark_edge_pipelines.sh \
@@ -136,10 +150,12 @@ cd ..
 * `-c` Classification device: `CPU` | `GPU` | `GPU.<idx>` | `NPU` (default: CPU)
 * `-i` Duration in seconds (default: 120)
 * `-t` CPU core type for pinning, e.g., `"ecore"` (optional)
-* `--concurrent` Enable concurrent GPU/NPU execution mode (optional)
+* `--concurrent` Enable concurrent GPU or NPU execution mode (optional)
+
+**Note:** Intel recommends the GPU or NPU for AI inference workloads.
+
+Step 5. Display Results:
 
-**Note:** GPU or NPU strongly recommended for AI inference workloads.
-#### Step 5: Display Results
 ```bash
 # Generate and view dashboard
 python3 html/generate_report.py
@@ -148,30 +164,27 @@ cd html && python3 -m http.server 8000  # Access at http://localhost:8000
 
 ### Output
 
-Results are saved to `results/` organized by execution mode:
+Results are saved to the `results/` folder, organized by execution mode:
 
-* `*.log` – Full GStreamer pipeline output (stdout/stderr)
-* `*.csv` – Performance metrics (FPS, stream density, configuration)
+* `*.log` – Full GStreamer pipeline output (stdout or stderr)
+* `*.csv` – Performance metrics (FPS, stream density, and configuration)
 
 ## Get Help or Contribute
 
 If you want to participate in the GitHub community for Edge Workloads and Benchmarks, you can
 contribute code, propose a design, download and try out a release, open an issue,
 benchmark application performance, and participate in
 [Discussions](https://github.com/open-edge-platform/edge-workloads-and-benchmarks/discussions).
+
 To learn more, check out the following resources:
 
 - [Open an issue](https://github.com/open-edge-platform/edge-workloads-and-benchmarks/issues)
 - [Submit a pull request](https://github.com/open-edge-platform/edge-workloads-and-benchmarks/pulls)
 - [Read the Contribution Guide](https://github.com/open-edge-platform/edge-microvisor-toolkit/blob/3.0/docs/developer-guide/emt-contribution.md)
 - [Report a security vulnerability](https://github.com/open-edge-platform/edge-workloads-and-benchmarks/blob/main/SECURITY.md)
 
-Before submitting a new report, check the existing issues to see if a similar one has not
-been filed already.
+Before submitting a new report, check the existing issues to see if a similar one has been filed.
 
-## License
+## Notices
 
 The **Edge Workload and Benchmarks** project is licensed under the [APACHE 2.0](./LICENSE) license.
-
----
-\* Other names and brands may be claimed as the property of others.

media-downloader/README.md

@@ -10,15 +10,15 @@ Downloads and prepares video files for Edge Workloads and Benchmarks pipelines.
 
 ## Overview
 
-1. Downloads two 4K test videos from Pexels
+1. Downloads two 4K test videos from Pexels platform.
    - `apple.mp4` - Single object per frame
    - `bears.mp4` - Two objects per frame
 
-2. Transcodes to H.265 format (1080p30, 2Mbps, no B-frames)
-   - Uses VA-API hardware acceleration via Docker
-   - Requires `/dev/dri` GPU/VA-API device access
+2. Transcodes to H.265 format (1080p30, 2Mbps, no B-frames).
+   - Uses video acceleration API (VA-API) hardware acceleration via Docker software.
+   - Requires `/dev/dri` GPU or VA-API device access.
 
-3. Loops each video 100x for long-duration benchmarks
+3. Loops each video 100 times for long-duration benchmarks.
 
 4. Saves looped videos to pipeline directories:
    - `pipelines/light/video/bears.h265`
@@ -27,5 +27,5 @@ Downloads and prepares video files for Edge Workloads and Benchmarks pipelines.
 
 ## Requirements
 
-- Docker with DLStreamer container (`intel/dlstreamer:latest`)
-- GPU with VA-API support (integrated/discrete GPU)
+- Docker software with Deep Learning Streamer (DL Streamer) container (`intel/dlstreamer:latest`).
+- GPU with VA-API support (integrated or discrete GPU).

model-conversion/README.md

@@ -1,8 +1,9 @@
 # Model Conversion and Quantization
 
-Download, convert, and quantize AI models for DLStreamer pipelines.
+Download, convert, and quantize AI models for Deep Learning Streamer (DL Streamer) pipelines.
 
 ### Model Collateral Matrix
+
 | Model Name   | Task           | Dimensions     | Dataset  | Source Model |
 |--------------|----------------|----------------|----------|--------------|
 | Yolo-v11n    | Detection      | 640x640 (INT8) | COCO     | [source](https://docs.ultralytics.com/models/yolo11/)   |
@@ -12,13 +13,14 @@ Download, convert, and quantize AI models for DLStreamer pipelines.
 | Mobilenet-V2 | Classification | 224x224 (INT8) | ImageNet | [source](https://pytorch.org/hub/pytorch_vision_mobilenet_v2/)   |
 
 ### Usage
+
 ```bash
 ./convert_models.sh
 ```
 
 ## ImageNet Accuracy Check for Classification Networks (Optional)
 
-CIFAR is used as a proxy dataset for classification network quantization. For classification accuracy validation, ImageNet is required.
+The CIFAR dataset is used as a proxy dataset for classification network quantization. For classification accuracy validation, the ImageNet dataset is required.
 
 ### ImageNet Dataset Setup
 
@@ -61,4 +63,4 @@ pipelines/
     └── classification/
         ├── resnet-v1-50-tf/INT8/
         └── mobilenet-v2-1.0-224-tf/INT8/
-```
+```

setup/drivers/README.md

@@ -1,6 +1,6 @@
 # Compute Driver Installation Scripts
 
-GPU and NPU compute drivers validated for Edge Workloads and Benchmarks performance reproducibility.
+Intel has validated the GPU and NPU compute drivers in terms of performance reproducibility, for Edge Workloads and Benchmarks.
 
 ## Usage
 
@@ -34,8 +34,8 @@ drivers/
 
 Downloaded packages are saved locally for offline reinstallation.
 
-## Integration with Main Prerequisite Script
+## Integration with the Main Prerequisite Script
 
 Driver installation is **optional** by default. The main `install_prerequisites.sh` script does not automatically install drivers to maintain system stability.
 
-Use these dedicated scripts or `install_prerequisites.sh` with the `--reinstall-gpu-driver=yes` / `--reinstall-npu-driver=yes` flags to install the compute drivers.
+Use these dedicated scripts or the `install_prerequisites.sh` script with the `--reinstall-gpu-driver=yes` / `--reinstall-npu-driver=yes` flags to install the compute drivers.