VC-1 Video

VC-1 is the video coding format standardized in SMPTE ST 421 with Simple, Main, and Advanced profiles.

VC-1 is the video coding format standardized in SMPTE ST 421. It is associated with the Windows Media video ecosystem and older movie collections. The useful source description includes its Simple, Main, or Advanced profile, rather than only the broader Windows Media name.

The standard reconstructs 8-bit 4:2:0 picture samples. Its coding tools determine how the compressed track turns into those samples.

Reconstruct an inter-coded region

The decoder selects the available reference information and applies the coded motion description to form a predictor. A correction block describes how the target region differs from that predictor.

VC-1 can use different transform shapes for inter-coded corrections:

Transform shape Region it represents
8 × 8 A square correction block
8 × 4 A horizontally extended sub-block
4 × 8 A vertically extended sub-block
4 × 4 A smaller square sub-block

Transform signaling and sub-block patterns tell the decoder which structure to apply. Different shapes can describe a localized correction more effectively than one fixed square basis for every region.

The coded coefficients are inverse-quantized and inverse-transformed. The decoder adds the resulting samples to the predictor. Applicable overlap-smoothing and loop-filter operations complete the specified reconstruction behavior.

As in other predictive codecs, the encoder chooses a representation, while the decoder follows its syntax. Quantization controls reconstruction precision; selecting a transform shape is a way to represent the correction, rather than a new display resolution.

Interpret the picture's scan structure

Advanced profile includes interlaced picture coding. The picture structure tells decoding and motion handling how its line sets relate.

For a native interlaced scene, alternating rows come from successive field times. Frame-shaped and field-shaped coding decisions have to preserve that relationship. A final progressive presentation then needs the appropriate field reconstruction.

For a film transfer, repeated or related field presentation can instead reflect a progressive source cadence. The picture's format and timing information help identify that case, with inverse telecine providing the corresponding recovery process.

This makes interlaced VC-1 more than a progressive picture with an “i” label: its decoder needs the relevant prediction and picture-structure operations.

Keep reconstruction and display interpretation together

ST 421 specifies coded reconstruction, while a complete player also interprets sample aspect, color, and timing information for presentation. Reconstructed samples with the wrong display shape are still being interpreted incorrectly.

In an Emby library inspection, record the actual VC-1 profile and scan structure along with its raster and rate. That identifies the reconstruction job before a playback route either presents the decoded source or supplies it to another encoder.

The format's standard reference is SMPTE ST 421:2013, VC-1 Compressed Video Bitstream Format and Decoding Process. It supplies the profile-specific syntax and reconstruction rules behind those observations.

See Also

References