PyNvVideoCodec 2.0.4__cp314-cp314-win_amd64.whl

samples/basic/simple_decode_sampling.py

@@ -0,0 +1,182 @@
+# This copyright notice applies to this file only
+#
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: MIT
+#
+# Permission is hereby granted, free of charge, to any person obtaining a
+# copy of this software and associated documentation files (the "Software"),
+# to deal in the Software without restriction, including without limitation
+# the rights to use, copy, modify, merge, publish, distribute, sublicense,
+# and/or sell copies of the Software, and to permit persons to whom the
+# Software is furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in
+# all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+# DEALINGS IN THE SOFTWARE.
+
+
+'''
+This sample demonstrates multi-file video decoding with frame sampling and tensor conversion.
+It shows how to:
+1. Process multiple video files efficiently
+2. Sample frames evenly across videos
+3. Convert video frames to PyTorch tensors
+4. Handle decoder reconfiguration
+
+Key Learning Outcomes:
+- Reconfigure decoder instance to process multiple video files
+- RGB color format output
+- PyTorch tensor conversion
+- CUDA device memory usage
+
+
+Usage:
+    python simple_decode_sampling.py video1.mp4 video2.mp4 [options]
+    python simple_decode_sampling.py /path/to/videos/*.mp4 [options]
+
+Example:
+    python simple_decode_sampling.py video1.mp4 video2.mp4 -g 0 -f 16
+
+Arguments:
+    video_files: One or more video files to process
+    -g/--gpu-id: GPU device ID (default: 0)
+    -f/--frames: Frames to sample per video (default: 8)
+
+Note: Frames are sampled evenly across each video's duration.
+'''
+
+import sys
+import torch
+import PyNvVideoCodec as nvc
+import numpy as np
+from typing import List
+from os.path import dirname, abspath
+
+# Add the parent directory to Python path in a way that works from any working directory
+current_dir = dirname(abspath(__file__))
+parent_dir = dirname(current_dir)
+if parent_dir not in sys.path:
+    sys.path.append(parent_dir)
+    
+# Import utilities for decoding
+from utils.Utils import convert_frames_to_torch
+from utils.decode_parser import parse_sampling_args
+    
+def process_multiple_files(video_files: List[str], gpu_id: int = 0, num_frames: int = 8, 
+                          need_scanned_stream_metadata: bool = False) -> None:
+    """
+    Process multiple video files using decoder reconfiguration.
+    
+    Args:
+        video_files: List of video file paths
+        gpu_id: GPU device ID to use
+        num_frames: Number of frames to decode per video
+        need_scanned_stream_metadata: Whether to scan stream for detailed metadata
+    """
+    if not video_files:
+        print("No video files provided")
+        return
+    
+    decoder = None
+    
+    try:
+        for i, video_file in enumerate(video_files):
+            # Convert Path to string if needed
+            video_path = str(video_file)
+            
+            print(f"\n{'='*60}")
+            print(f"Processing file {i+1}/{len(video_files)}: {video_path}")
+            print(f"{'='*60}")
+            
+            if i == 0:
+                # Create decoder from first file
+                print(f"Creating decoder for: {video_path}")
+                decoder = nvc.SimpleDecoder(
+                    video_path,
+                    gpu_id=gpu_id,
+                    use_device_memory=True,
+                    need_scanned_stream_metadata=need_scanned_stream_metadata,
+                    output_color_type=nvc.OutputColorType.RGB,  # RGB format
+                    max_width=2048,
+                    max_height=2048
+                )
+                print(f"Decoder created successfully. Total frames: {len(decoder)}")
+            else:
+                # Reconfigure decoder for subsequent files
+                print(f"Reconfiguring decoder for: {video_path}")
+                torch.cuda.current_stream().synchronize()
+                decoder.reconfigure_decoder(video_path)
+                print(f"Decoder reconfigured successfully. Total frames: {len(decoder)}")
+            
+            # Get total frames and calculate indices
+            total_frames = len(decoder)
+            
+            if total_frames == 0:
+                print(f"Warning: No frames found in video: {video_file}")
+                continue
+            
+            # Sample frames evenly across the video
+            frame_indices = np.linspace(0, total_frames-1, num_frames, dtype=int).tolist()
+            
+            print(f"Decoding {num_frames} frames from indices: {frame_indices}")
+            
+            # Get batch of frames by indices
+            decoded_frames = decoder.get_batch_frames_by_index(frame_indices)
+            
+            # Convert to torch tensor
+            frames_tensor = convert_frames_to_torch(decoded_frames)
+            
+            print(f"Decoded frames shape: {frames_tensor.shape}")
+            
+            # Clean up frames to free memory
+            del frames_tensor
+            del decoded_frames
+            
+        print(f"\n{'='*60}")
+        print("All files processed successfully!")
+        print(f"{'='*60}")
+        
+    except Exception as e:
+        print(f"Error processing files: {e}")
+        raise
+    finally:
+        # Clean up decoder
+        if decoder is not None:
+            del decoder
+def main():
+    """Main application entry point"""
+    try:
+        args, valid_files = parse_sampling_args()
+        
+        # Display configuration
+        print("=" * 60)
+        print("VIDEO SAMPLING CONFIGURATION")
+        print("=" * 60)
+        print(f"GPU: {args.gpu_id} ({torch.cuda.get_device_name(args.gpu_id)})")
+        print(f"Input Files: {len(valid_files)} files found")
+        print(f"Frames per Video: {args.frames}")
+        print(f"Need Scanned Stream Metadata: {args.need_scanned_stream_metadata}")
+        print("=" * 60)
+        
+        # Process all files
+        process_multiple_files(valid_files, args.gpu_id, args.frames, args.need_scanned_stream_metadata)
+        
+        print("=" * 60)
+        print("SAMPLING COMPLETE")
+        print("=" * 60)
+        
+    except KeyboardInterrupt:
+        print("\nApplication interrupted by user")
+        sys.exit(0)
+    except Exception as e:
+        print(f"Application error: {e}")
+        sys.exit(1)
+if __name__ == "__main__":
+    main() 

samples/basic/transcode_config.json

@@ -0,0 +1,6 @@
+{
+    "codec" : "h264",
+    "preset" : "P3",
+    "tuning_info" : "high_quality",
+    "bf" : "3"
+}