Inference Scripts

Command-line interfaces for image and video upsampling

inference_realesrgan.py

Command-line script for upsampling images using Real-ESRGAN models.

Usage

python inference_realesrgan.py -i inputs -o results -n RealESRGAN_x4plus

Arguments

-i, --input · str · default: "inputs"

Input image or folder path

-n, --model_name · str · default: "RealESRGAN_x4plus"

Model name to use for inference

Available models:

• RealESRGAN_x4plus - General 4x upsampling
• RealESRNet_x4plus - 4x upsampling without GAN
• RealESRGAN_x4plus_anime_6B - Anime images 4x
• RealESRGAN_x2plus - General 2x upsampling
• realesr-animevideov3 - Anime video 4x
• realesr-general-x4v3 - General purpose 4x with denoise control

-o, --output · str · default: "results"

Output folder path

-dn, --denoise_strength · float · default: "0.5"

Denoise strength. Range: 0 (weak denoise, keep noise) to 1 (strong denoise ability)

Only used for the realesr-general-x4v3 model

-s, --outscale · float · default: "4"

The final upsampling scale of the image

--model_path · str · default: "None"

Optional custom model path. Usually not needed as models are downloaded automatically

--suffix · str · default: "out"

Suffix of the restored image filename

-t, --tile · int · default: "0"

Tile size for processing. 0 means no tiling. Use tiling to avoid out-of-memory errors with large images

--tile_pad · int · default: "10"

Tile padding size to reduce border artifacts

--pre_pad · int · default: "0"

Pre-padding size at each border

--face_enhance · bool

Use GFPGAN to enhance faces in the image (flag, no value needed)

--fp32 · bool

Use fp32 (full) precision during inference. Default is fp16 (half precision)

--alpha_upsampler · str · default: "realesrgan"

The upsampler for alpha channels (transparency)

Options: realesrgan | bicubic

--ext · str · default: "auto"

Output image extension

Options: auto | jpg | png

auto uses the same extension as the input

-g, --gpu-id · int · default: "None"

GPU device to use. Can be 0, 1, 2, etc. for multi-GPU systems

Examples

Basic Usage

# Upscale a single image
python inference_realesrgan.py -i input.jpg -o output_folder -n RealESRGAN_x4plus

Memory Management: If you encounter CUDA out of memory errors, try using the --tile option with a smaller tile size (e.g., -t 200 or -t 400). Tiling processes the image in smaller chunks at the cost of slightly slower processing.

File Formats: RGBA images (with transparency) are automatically saved as PNG regardless of the --ext setting to preserve the alpha channel.

inference_realesrgan_video.py

Command-line script for upsampling videos using Real-ESRGAN models. Optimized for anime videos.

Usage

python inference_realesrgan_video.py -i video.mp4 -o results -n realesr-animevideov3

Arguments

-i, --input · str · default: "inputs"

Input video, image, or folder path

-n, --model_name · str · default: "realesr-animevideov3"

Model name to use for inference

Available models:

• realesr-animevideov3 - Optimized for anime videos (default)
• RealESRGAN_x4plus_anime_6B - Anime images/videos 4x
• RealESRGAN_x4plus - General 4x upsampling
• RealESRNet_x4plus - 4x upsampling without GAN
• RealESRGAN_x2plus - General 2x upsampling
• realesr-general-x4v3 - General purpose 4x with denoise control

-o, --output · str · default: "results"

Output folder path

-dn, --denoise_strength · float · default: "0.5"

Denoise strength. Range: 0 (weak denoise, keep noise) to 1 (strong denoise ability)

Only used for the realesr-general-x4v3 model

-s, --outscale · float · default: "4"

The final upsampling scale of the video

--suffix · str · default: "out"

Suffix of the restored video filename

-t, --tile · int · default: "0"

Tile size for processing. 0 means no tiling. Use tiling to avoid out-of-memory errors

--tile_pad · int · default: "10"

Tile padding size to reduce border artifacts

--pre_pad · int · default: "0"

Pre-padding size at each border

--face_enhance · bool

Use GFPGAN to enhance faces in the video (flag, no value needed)

Note: Automatically disabled for anime models

--fp32 · bool

Use fp32 (full) precision during inference. Default is fp16 (half precision)

--fps · float · default: "None"

FPS of the output video. If not specified, uses the input video’s FPS

--ffmpeg_bin · str · default: "ffmpeg"

Path to the ffmpeg binary

--extract_frame_first · bool

Extract frames to disk before processing (flag, no value needed). Can be useful for certain workflows

--num_process_per_gpu · int · default: "1"

Number of processes to spawn per GPU for parallel processing

--alpha_upsampler · str · default: "realesrgan"

The upsampler for alpha channels (transparency)

Options: realesrgan | bicubic

--ext · str · default: "auto"

Image extension when processing image folders

Options: auto | jpg | png

Examples

Basic Video Upscaling

# Upscale anime video with default settings
python inference_realesrgan_video.py -i anime_video.mp4 -o results/

Performance Warning: If you are generating videos larger than 4K resolution, processing will be very slow due to I/O speed limitations. It is highly recommended to decrease the --outscale value.

Multi-GPU Processing: The script automatically detects available GPUs and can process video segments in parallel. Use --num_process_per_gpu to control how many processes run per GPU. For example, with 2 GPUs and --num_process_per_gpu 2, a total of 4 processes will run in parallel.

FLV Format: If the input is a .flv file, the script automatically converts it to .mp4 using ffmpeg before processing.

Audio Preservation: The script automatically preserves the audio track from the input video in the output video.

Helper Classes

The video inference script includes two helper classes:

Reader

Handles reading frames from videos, images, or folders. Supports streaming from video files using ffmpeg.

Methods:

• get_resolution() - Returns (height, width) of the input
• get_fps() - Returns the FPS of the video
• get_audio() - Returns the audio stream
• get_frame() - Returns the next frame
• close() - Closes the stream reader

Writer

Handles writing frames to output video using ffmpeg with H.264 encoding.

Methods:

• write_frame(frame) - Writes a frame to the output video
• close() - Finalizes and closes the output video