Back to Search Document details
13th Meeting: Marrakech, January 2019 2019-01-10 11:03
CE4: Non-scaling STMVP (Test 4.2.1)
Authors: Tianyang Zhou Sharp Tomohiro Ikai Sharp
Abstract
The draft text for Spatial-temporal merge mode is provided based on Versatile Video Coding Draft 3 (JVET-L1001).
Meeting Notes
JVET-M0059 CE4: Non-scaling STMVP (Test 4.2.1) [T. Zhou, T. Ikai (Sharp)]

The proposed method derives an averaging candidate as STMVP candidate using two spatial merge candidates and one collocated merge candidate.

i = 0
if( availableFlagA1 )
mergeCandList[ i++ ] = A1
if( availableFlagB1 )
mergeCandList[ i++ ] = B1
if( availableFlagB0 )
mergeCandList[ i++ ] = B0 (8-118)
if( availableFlagA0 )
mergeCandList[ i++ ] = A0
if( availableFlagSbCol )
mergeCandList[ i++ ] = STMVP
if( availableFlagB2 )
mergeCandList[ i++ ] = B2
if( availableFlagCol )
mergeCandList[ i++ ] = Col

For the spatial candidates, the first and second candidates in the current merge candidate list is used.

For the temporal candidate, the same position as VTM / HEVC collocated position is used.

If three candidates, of which the reference are equal to zero, are available, the following apply.

mvLX[0] = (mvLX_A[0] * 3 + mvLX_L[0] * 3 + mvLX_C[0] * 2 ) / 8

mvLX[1] = (mvLX_A[1] * 3 + mvLX_L[1] * 3 + mvLX_C[1] * 2 ) / 8

If two motion information, of which the reference are equal to zero, is available, the following apply

mvLX[0] = (mvLX_A[0] + mvLX_C[0] ) / 2

mvLX[1] = (mvLX_A[1] + mvLX_C[1] ) / 2

or

mvLX[0] = (mvLX_B[0] + mvLX_C[0] ) / 2

mvLX[1] = (mvLX_B[1] + mvLX_C[1] ) / 2

Note: If the temporal candidate is unavailable, the STMVP mode is off.

References:
JVET-L1001
Cited:
JVET-M0024
ARCHIVE v2 :
CE4.2.1.a_STMVP.patch CE4.2.1.a_tester.xlsm CE4.2.1_additional_test_results.xlsm JVET-M0059_v2.doc WD_STMVP_M0059_CE4.2.1a.docx ~$ET-M0059.doc
Decisions
Note: If the temporal candidate is unavailable, the STMVP mode is off.
Citation