Processing digital video directly in the compressed domain has many advantages in terms of storage efficiency, speed, and video quality. We have developed a compressed video editing and parsing system (CVEPS) with advanced video indexing and manipulation functions. The video parsing tools support automatic extraction of key visual features, e.g., scene cuts, transitional effects, camera operations (zoom/pan), shape and trajectories of prominent moving objects. These visual features are used for efficient video indexing, retrieval and browsing. The editing tools allow users to perform useful video compositing functions and special visual effects typically seen in video production studios. We contrast our compressed-domain approach with traditional decode-process-reencode approach with quantitative and/or qualitative performance comparison. We also present a client-server network based CVEPS implementation. KEYWORDS Compressed domain video manipulation, client-server network based video...