ISO/IEC 23008-12:2025/Amd 2:2026

International
Standard
ISO/IEC 23008-12
Third edition
Information technology — High
2025-07
efficiency coding and media
delivery in heterogeneous
AMENDMENT 2
environments —
2026-06
Part 12:
Image File Format
AMENDMENT 2: Low-overhead image
file format
Technologies de l'information — Codage à haute efficacité et
livraison des medias dans des environnements hétérogènes —
Partie 12: Format de fichier d'image
AMENDEMENT 2: Format de fichier image à entête optimisé
Reference number
ISO/IEC 23008-12:2025/Amd. 2:2026(en) © ISO/IEC 2026

ISO/IEC 23008-12:2025/Amd. 2:2026(en)
© ISO/IEC 2026
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
© ISO/IEC 2026 – All rights reserved
ii
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical activity.
ISO and IEC technical committees collaborate in fields of mutual interest. Other international organizations,
governmental and non-governmental, in liaison with ISO and IEC, also take part in the work.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of document should be noted. This document was drafted in accordance with the editorial rules of the ISO/
IEC Directives, Part 2 (see www.iso.org/directives or www.iec.ch/members_experts/refdocs).
ISO and IEC draw attention to the possibility that the implementation of this document may involve the
use of (a) patent(s). ISO and IEC take no position concerning the evidence, validity or applicability of any
claimed patent rights in respect thereof. As of the date of publication of this document, ISO and IEC had
received notice of (a) patent(s) which may be required to implement this document. However, implementers
are cautioned that this may not represent the latest information, which may be obtained from the patent
database available at www.iso.org/patents and https://patents.iec.ch. ISO and IEC shall not be held
responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO’s adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT) see www.iso.org/iso/foreword.html.
In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology, SC 29,
Coding of audio, picture, multimedia and hypermedia information.
A list of all parts in the ISO/IEC 23008 series can be found on the ISO and IEC websites.
Any feedback or questions on this document should be directed to the user’s national standards
body. A complete listing of these bodies can be found at www.iso.org/members.html and
www.iec.ch/national-committees.

© ISO/IEC 2026 – All rights reserved
iii
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
Information technology — High efficiency coding and media
delivery in heterogeneous environments —
Part 12:
Image File Format
AMENDMENT 2: Low-overhead image file format

Introduction
Add the following paragraphs at the end of the Introduction:
Annex O specifies the Low-overhead Image File Format suitable for use cases requiring small and
simple images. This format reduces the overhead relative to the structure-data describing the image
or metadata payloads for small image files compared to the regular Image File Format. Annex O also
specifies the requirements and brands for this format, as well as the procedure for expanding it to the
regular Image File Format.
Annex P specifies the MIME type registration for a single image for the Low-overhead Image File Format-
specific brand.
3.1.11
Add the following note to entry after the definition:
Note 1 to entry The term “track” is defined in ISO/IEC 14496-12.

3.1.58
Add the following new term and definition:
3.1.58 tiled image item
image item of type 'tili' constructed using uniform tiled subregions, allowing direct access to individual
tiles
6.1
Add the following NOTE after the first paragraph:
NOTE For small and simple files, a pre-processed version of the MetaBox called the MinimizedImageBox, as
defined in Annex O, can be used. A file containing a MinimizedImageBox has the same compliance and rendering
requirements as an Image File Format file because the MinimizedImageBox is expanded into the equivalent MetaBox
and MediaDataBox as described in detail in Annex O.4. Once expanded, the rest of Clause 6 applies.

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
6.5.1
Number the existing NOTE as NOTE 2.
Replace the third and fourth paragraphs with:
The semantics of the descriptive properties specified in 6.5 are specified for the image before the
transformations, if any, are applied.
NOTE 1 It is uncertain if readers would be able to correctly interpret descriptive properties that follow the
first transformative property or the first unrecognized essential property, whichever is earlier, in the sequence
associating properties with an item, because those descriptive properties possibly describe the output image after
the transformation(s). In previous versions of this document, readers had to allow and ignore descriptive properties
following the first transformative or unrecognized property, whichever is earlier, in the sequence associating
properties with an item.
Writers should arrange the descriptive properties prior to any transformative property in the sequence
associating properties with an item. Writers should arrange the descriptive properties specified in 6.5 prior
to any other properties in the sequence associating properties with an item.

6.5.6.1
Replace:
The PixelInformationProperty descriptive item property indicates the number and bit depth of colour
components in the reconstructed image of the associated image item.
with:
The PixelInformationProperty descriptive item property indicates the number and bit depth of colour and
alpha/depth components, if present, in the reconstructed image of the associated image item. If px_flags & 1
!= 0, the PixelInformationProperty also indicates content, component format and subsampling information
per channel.
6.5.6.2
Replace the existing text with the following:

aligned(8) class PixelInformationProperty
extends ItemFullProperty('pixi', version = 0, px_flags){
unsigned int(8) num_channels;
for (i=0; i unsigned int(8) bits_per_channel;
}
if((px_flags & 1) != 0) {
for (i=0; i unsigned int(3) channel_idc;
unsigned int(1) reserved = 0;
unsigned int(2) component_format;
unsigned int(1) subsampling_flag;
unsigned int(1) channel_label_flag;
if(subsampling_flag) {
unsigned int(4) subsampling_type;
unsigned int(4) subsampling_location;
}
if(channel_label_flag) {
utf8string channel_label;
}
}
}
}
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)

6.5.6.3
Replace:
bits_per_channel: This field indicates the bits per channel for the pixels of the reconstructed image of the
associated image item.
With:
bits_per_channel: This field indicates the bits per channel for the pixels of the reconstructed image of the
associated image item. The value of this field shall not be 0.

6.5.6.3
Add the following text to the end of the subclause:
px_flags&1: If equal to 1, indicates that the channel_idc, component_format, subsampling_flag, and
channel_label_flag fields are present. If equal to 0, indicates that the channel_idc, component_format,
subsampling_flag and channel_label_flag fields are not present.
channel_idc: This field indicates the contents of the channel as specified
in Table 13. At most one channel shall have a channel_idc of 5.
Table 13 — channel_idc values and their meaning
Value of channel_idc Mapping (depending on the 'colr' box)
0 Unused
1 Unspecified
2 First colour channel (e.g. monochrome, Y, R, C)
3 Second colour channel (e.g. U, Cb, G, M)
4 Third colour channel (e.g. V, Cr, B, Y)
5 Alpha
6 Depth
7 Fourth colour channel (e.g. K)
component_format: This field indicates the data type of the channel as defined by the component_format
values in ISO/IEC 23001-17 where component_bit_depth is considered to be equal to bits_per_channel.
subsampling_flag: If equal to 1, indicates that the subsampling_type and subsampling_location fields are
present. If equal to 0, indicates that the subsampling_type and subsampling_location fields are not
present.
channel_label_flag: If equal to 1, indicates the presence of the channel_label field. If equal to 0, indicates
the channel_label field is not present.
subsampling_type: This field indicates the subsampling type as specified by Table 14.
subsampling_location: This field indicates the subsampling sample location as specified by Table 14.
channel_label: The human readable description of the channel.

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
Table 14 — Channel subsampling and sample position
Position of the centre of the top-
left chroma sample relative to the
Channel
Value of Value of subsampling_
centre of the top-left luma sample
subsampling_type location
subsampling
in units of luma samples
Horizontal (x) Vertical (y)
0 0 or 1 or 2 or 3 or 4 or 5 none (4:4:4) 0 0
0 or 2 or 4 subsampled by a 0
1 factor 2 horizontally 0
1 or 3 or 5 0.5
(4:2:2)
0 0 0.5
1 0.5 0.5
subsampled both
2 0 0
horizontally and
vertically by a factor
3 0.5 0
2 (4:2:0)
4 0 1
5 0.5 1
0 or 2 or 4 subsampled by a 0
3 factor 4 horizontally 0
1 or 3 or 5 1.5
(4:1:1)
0 or 1 0.5
subsampled by a
4 2 or 3 factor 2 vertically 0 0
(4:4:0)
4 or 5 1
Other values Reserved
6.5.8.2
Change the syntax to:
aligned(8) class AuxiliaryTypeProperty
extends ItemFullProperty('auxC', version = 0, flags) {
utf8string aux_type;
unsigned int(8) aux_subtype[];
// until the end of the box, the semantics depend on the aux_type value
}
6.5.10.3, 6.5.13.1, 6.6.2.3.1
Replace all occurrences of “*” with “×”.

6.6.2.3.1
Set all occurrences of “tile_width” and “tile_height” in italic font.

6.6.2.5.3
Change the table in subclause 6.6.2.5.3 of ISO/IEC 23008-12:2025/Amd 1:2025 to:

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
channel_id Mapping (depending on the 'colr' box)
or
packed_channel_id
0 Unused
1 Unspecified
2 Y R C
3 Cb G M
4 Cr B Y
5 Alpha
6 Depth
7 Reserved Reserved K
8-255 Reserved for future use.

Add the following in subclause 6.6.2.5.3 of ISO/IEC 23008-12:2025/Amd 1:2025, after channel_id:
channel_id, when present, shall not be equal to 0 (Unused).
The regions indicated by all instances of channel_id and packed_channel_id not equal to 0 present in the
current colour format enhancement derived image item, are used to generate a derived image based on the
values of the indicated channel_id and packed_channel_id and any associated colour information.
If is_packed_flag is equal to 1, then the picture is first partitioned into (num_cols_minus1+1)×(num_rows_
minus1+1) regions as follows:
The width and height of each region is computed as follows:
RegionWidthInLumaSamples = image_width / num_cols_minus1
RegionHeightInLumaSamples = image_height / num_rows_minus1
A region R[j][k] for any value of j in the range of 0 to num_rows_minus1, inclusive, and k in the range of 0
to num_cols_minus1, inclusive, is associated with the luma area starting from the horizontal and vertical
coordinates OrigX[j][k] and OrigY[j][k] that are computed as follows:
OrigX[j][k] = (k×RegionWidthInLumaSamples)+(hor_guard_band_mul2×2)
OrigY[j][k] = (j×RegionHeightInLumaSamples)+(ver_guard_band_mul2×2)
and end at horizontal and vertical coordinates EndX[j][k] and EndY[j][k] that are computed as follows:
EndX[j][k]= ((k+1)×RegionWidthInLumaSamples)-(hor_guard_band_mul2×2+1)
EndY[j][k]= ((j+1)×RegionHeightInLumaSamples)-(ver_guard_band_mul2×2+1)
Only the luma samples associated with any indicated regions, as specified above, are considered for the
generation of a derived image and all other luma and chroma samples are ignored.

6.8.12.1
Change the following sentence from subclause 6.8.12.1 of ISO/IEC 23008-12:2025/Amd 1:2025:
When the flag tile_info_present_flag is not set, the tile information of a layer of the image pyramid is
derived depending on the image item as described in Tables 2 to 4 below.
To:
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
When the flag tile_info_present_flag is not set, the tile information of a layer of the image pyramid is
derived depending on the image item as described in Tables 2 to 5 below.

Add the following Table 15 after Table 4 in subclause 6.8.12.1 of ISO/IEC 23008-12:2025/Amd 1:2025:
Table 15 — Tile information based on Tiled image item 'tili'
ImagePyramidEntityGroup tile Tiled image item 'tili'
information
tile_width
tileWidth
tile_height
tileHeight
tileColumns ceil(ispe.image_width/tile_width)
tileRows ceil(ispe.image_height/tile_height)

6.11
Add the following new subclause after subclause 6.10.4.1.3:
6.11  Tiled image items
6.11.1  General
A tiled image item is constructed of uniform, independently coded tiles arranged in rows, columns, and
optionally extra dimensions, to form a rectangular image or n-dimensional hyperrectangle. The tiles are
identical in size, format, coding, and makeup and may be compressed or uncompressed.
6.11.2  Tiled image item
An image item of type 'tili' is a tiled image, with each tile coded independently from other tiles. Input tiles
may be stored either in separate external files or in a contiguous range of addressing space to support byte
range addressing and retrieval of individual tiles with a single read.
NOTE 1 The image coding method is defined by a writer using a valid image codec 4CC.
NOTE 2 As opposed to a 'grid' image item, where the declaration and addressing of tiles occurs in the file-scoped
MetaBox, a 'tili' image item with contiguous range of addressing space has a single declaration parameter in the
file-scoped MetaBox and an associated addressing table, with offsets and extents for each tile, stored with the image
tiles, typically in a media data box. This has the advantage that the required file ranges of the addressing table, which
can be large for terapixel images, can also be loaded on-demand.
The tiled image item ('tili') shall be associated with TiledImageConfigurationProperty ('tilC') and
ImageSpatialExtentsProperty which carries the width and height of the overall tiled image.
NOTE 3 The tiled image item ('tili') allows for storing YUV imagery as a single coded tile object. Conversely, each
band in a tile can be stored separately, allowing for direct access to each component in a tile separately. This results in
a 3D arrangement of tiles within the image. This is useful for multi and hyperspectral imagery.
NOTE 4 A tiled image item ('tili') can be included within an ImagePyramidEntityGroup. To create a full
pyramid, progressively binned tiled image items may be included. When building a multi-resolution pyramid, different
image coding methods can be used for each layer.
The location of input tiles to the tiled image item is identified by the corresponding DataEntryTiledItemBox
in the DataReferenceBox which is mapped to the tiled image item through data_reference_index in the
ItemLocationBox.
The DataEntryTiledItemBox shall contain URLs to the input tiles stored in separate external files or shall
contain the size and offset to the TiledImageOffsetTable to support byte range addressing and retrieval of
individual tiles.
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
If the input tiles are stored in separate external files, then each external file shall contain only one input tile
as an image item and be a HEIF compliant file. The external files shall not contain any entity grouping. The
input tile in an external file may contain item properties associated with them. The handler_type of the tiled
image item shall be equal to the handler_type of the input tile in the external file.
When DataEntryTiledItemBox.external_tiles_urls is equal to 0, all the necessary item properties for
the input tiles of the tiled image item shall be present in the ItemPropertyBox and are associated to the
tiles of the tiled image item through the TileItemPropertyAssociationBox. When DataEntryTiledItemBox.
external_tiles_urls is different from 0, all the necessary item properties for an input tile stored in an
external file are present in the external file.
When the overall image dimensions are not an even multiple of the image tile size, the rows are padded on
the right to complete the last tile in each row of tiles, and the columns are padded on the bottom to complete
the last tile in each column of tiles. The width and height parameters in the ImageSpatialExtentsProperty
are set to the size of the image containing valid image content, effectively achieving a crop of the padded
boundary area.
When the input tiles are stored in separate external files, the coding format of the tiled image item is
given by the item_type of the image item in the external file. When the input tiles are stored in the same
file as the tiled image item, the coding format of the tiled image item is given by tile_item_type in the
TiledImageConfigurationProperty associated with the tiled image item.
6.11.3  Tiled image configuration property
6.11.3.1  Definition
'tilC'
Box type:
Property type: Descriptive item property
ItemPropertyContainerBox
Container:
Mandatory (per item): Yes, for image items of type 'tili'
Quantity (per item): One
The TiledImageConfigurationProperty specifies parameters associated with a tiled image item ('tili')
(see Section 6.12). These parameters include the tile resolution, and the image item type used to code and
store individual tile content. Configuration information also includes the number and size of additional
dimensions when coding n-dimensional hyperrectangles. This includes support for the coding of multi and
hyperspectral imagery where each band in a tili tile region is separately retrievable.
The TiledImageConfigurationProperty consists of TileItemPropertyAssociationBox which associates tiles
to item properties in the ItemPropertiesBox.
6.11.3.2  Syntax
aligned(8) class TileItemPropertyAssociationBox
extends FullBox('tipa', version=0, flags=0) {
unsigned int(8) association_count;
for (j=0; j bit(1) essential;
if (flags & 1)
unsigned int(15) property_index;
else
unsigned int(7) property_index;
}
}
aligned(8) class TiledImageConfigurationProperty
extends ItemFullProperty('tilC', version=0, flags=0) {
unsigned int(32) tile_width;
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
unsigned int(32) tile_height;
unsigned int(8) number_of_extra_dimensions;
for (int i=0; i unsigned int(32) dimension_size[i];
}
if (DataEntryTiledItemBox.external_tiles_urls==0) {
unsigned int(32) tile_item_type;
TileItemPropertyAssociationBox tile_association;
}
}
6.11.3.3  Semantics
tile_width, tile_height shall be set to the size of a single tile width and height. All tiles have the same
size. Tiles at the right or bottom border of the overall image may include padding when the tile width
and or height are not integer multiples of the overall tiled image item width or height. In this case, the
ImageSpatialExtentsProperty is set to the boundary of the true image width and height to achieve a
crop of the padded area.
tile_item_type specifies the image item type used for all the individual tile images. In a tiled image item,
each tile is coded separately so it can be extracted and decoded independently. tile_item_type shall
be set to a valid four-character code for a coded image item (e.g., 'hvc1' for h265 compression, 'j2k1'
for JPEG2000, or 'unci' for uncompressed). When required by the image item type, all necessary
image properties shall be associated using the TileItemPropertyAssociationBox. Certain codecs
(jpg, etc.) may not require any configuration properties in such a case the association_count in the
TileItemPropertyAssociationBox is set to 0.
number_of_extra_dimensions specifies the number of extra dimensions if the image resembles a (number_
of_extra_dimensions+2)-dimensional hyperrectangle. For a 2D image, number_of_extra_dimensions
shall be 0.
For a single layer 2D image (e.g., YUV or monochrome), tiles are indexed with a uniform ordering of the form
[y][x], where [x] is the width parameter, and [y] is the height parameter. For a multi-dimensional image
where components are stored separately, the tiles are indexed with a uniform ordering based on the
number_of_extra_dimensions, such as when equal to 0, the form is [y][x], when equal to 1, the form is [z]
[y][x], when equal to 2, the form is [a][z][y][x], and so on.
dimension_size[i] specifies the size of dimension i+2 of the n-dimensional hyperrectangle.
NOTE The size of the first two dimensions are the image_width and image_height specified in the
ImageSpatialExtentsProperty of the 'tili' item.
association_count indicates the number of item properties associated with the tiles of the tiled image item.
essential when set to 1 indicates that the associated property is essential to the tiles of the tiled image
item, otherwise it is non-essential.
property_index is either 0 indicating that no property is associated (the essential indicator shall also be
0), or is the 1-based index (counting all boxes, including FreeSpace boxes) of the associated property
box in the ItemPropertyContainerBox contained in the ItemPropertiesBox. property_index shall not be
greater than the number of boxes contained in the ItemPropertyContainerBox. The (flags & 1) value in
TileItemPropertyAssociationBox should be equal to 0 unless there are more than 127 properties in the
ItemPropertyContainerBox.
6.11.4  Tiled image item data
The payload of a tiled image item ('tili') consists of tiles of the item when the external_tiles_urls in the
associated DataEntryTiledItemBox is set to 0.
6.11.5  Data entry tiled item box
6.11.5.1  Definition
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
'deti'
Box Type:
DataReferenceBox
Container:
Mandatory: No.
Quantity: Zero or more.
The DataEntryTiledItemBox identifies the location of each tile of a tiled image item ('tili'). The
DataEntryTiledItemBox either contains URLs to the input tiles stored in separate external files or contains
the size and offset of the TiledImageOffsetTable to support byte range addressing and retrieval of individual
tiles.
The TiledImageOffsetTable contains offset pointers and size information for each tile in the tiled image
item. The TiledImageOffsetTable is stored in the MediaDataBox of the file.
6.11.5.2  Syntax
aligned(8) class TiledImageOffsetTable{
for (int i=0; i < NumTiles; i++) {
unsigned int(DataEntryTiledItemBox.offset_field_length)
tile_start_offset[i];
if (DataEntryTiledItemBox.size_field_length > 0) {
unsigned int(DataEntryTiledItemBox.size_field_length) tile_size[i];
}
aligned(8) class DataEntryTiledItemBox (bit(24) flags)
extends DataEntryBaseBox('deti', flags)
{
switch (flags & 0x03) {
case 0:
offset_field_length = 32;
break;
case 1:
offset_field_length = 40;
break;
case 2:
offset_field_length = 48;
break;
case 3:
offset_field_length = 64;
break;
}
switch ((flags>>2) & 0x03) {
case 0:
size_field_length = 0;
break;
case 1:
size_field_length = 24;
break;
case 2:
size_field_length = 32;
break;
case 3:
size_field_length = 64;
break;
}
sequential_order = ((flags>>4) & 0x01);
switch ((flags>>5) & 0x03) {
case 0:
input_items_size_index = 8;
break;
case 1:
input_items_size_index = 16;
break;
case 2:
input_items_size_index = 32;
© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
break;
case 3:
input_items_size_index = 64;
break;
}
external_tiles_urls = ((flags>>7) & 0x01);
unsigned int(input_items_size_index) no_of_input_items;
if(external_tiles_urls){
bit(7) reserved = 0;
bit(1) directory_ID_flag;
if(directory_ID_flag){
unsigned int(16) directoryIDstart;

unsigned int(16) directoryIDend;

}
unsigned int(64) tileIDstart;
utf8string baseurl;
utf8string urlextension;
utf8string tileitemrequesttemplate;
}
else{
unsigned int(DataEntryTiledItemBox.offset_field_length)
tile_offset_table_start_offset;
unsigned int(32)
tile_offset_table_size;
}
}
6.11.5.3  Semantics
offset_field_length defines the number of bits used to store the offset to the TiledImageOffsetTable and
the offset to the image data of a specific tile in the TiledImageOffsetTable.
size_field_length defines the number of bits used to store the length of the TiledImageOffsetTable and
the length of the image data of a specific tile in the TiledImageOffsetTable.
sequential_order If equal to 1, indicates that the compressed image tile data is stored consecutively in
sequential order.
input_items_size_index specifies the size of the parameters no_of_input_items in bytes. With value 0
indicating size is of 1 byte up to the value 3 indicating the size to be 8 bytes.
external_tile_urls If equal to 1, indicates that the tiles are stored in external files indicated by the URLs.
If equal to 0, indicates that the tiles are stored within the same file.
The parameter no_of_input_items in DataEntryTiledItemBox shall be equal to:

TileColumns = (ImageSpatialExtentsProperty.image_width +
TiledImageConfigurationProperty.tile_width-1)/
TiledImageConfigurationProperty.tile_width;
TileRows  = (ImageSpatialExtentsProperty.image_height +
TiledImageConfigurationProperty.tile_height-1)/
TiledImageConfigurationProperty.tile_height;
no_of_input_items = TileColumns × TileRows
for (i=0; i< TiledImageConfigurationProperty.number_of_extra_dimensions; i++)
{
no_of_input_items = no_of_input_items ×
TiledImageConfigurationProperty.dimension_size[i];
}
directory_ID_flag If equal to 1, indicates the presence of directory IDs to be used for URL construction. If
equal to 0, indicates directory IDs are not used in URL construction.

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
directoryIDstart indicates the ID of the first directory to be used in URL construction.
directoryIDend indicates the ID of the last directory to be used in URL construction.
The number of tiles in each directory is equal to NumTiles/(directoryIDend-directoryIDstart). When the
number of tiles in each directory is not an integer factor of NumTiles then the last directory contains the
remaining tiles.
tileIDstart indicates the ID of the first tile in the associated tiled image item. The tile IDs start with
tileIDstart value until NumTiles + tileIDstart with increments of one in the order it is mapped in the tiled
image item (row-major order).
baseurl contains the base URL for the tiles.
urlextension contains URL extensions which is used in URL
tileitemrequesttemplate contains the template which is used in URL construction
base URL http://cdn.example.com/pictures/134532/image/
URL extension Representation1
Directory ID Start
Directory ID End 25
Tile Item Request template $tileID$.heif
Tile ID Start 1000
URL http://cdn.example.com/pictures/134532/image/
Representation1/10/1000.heif
In the example above, assuming that the first tile is selected, the URL constructed results in http:// cdn
.example .com/ pictures/ 134532/ image/ Representation1/ 10/ 1000 .heif
The URL is constructed by concatenating the baseurl and the urlextension and the ‘/’ character and the
directory ID to which the tile belongs to followed by the value of the tileitemrequesttemplate, where the
tileID is replaced by the actual value of the tile starting from the value given in tileIDstart.
tile_offset_table_start_offset points to the start of the TiledImageOffsetTable. The position is given
relative to the referenced data in the ItemLocationBox associated to the tiled image item.
tile_offset_table_size indicates the size in bytes of the TiledImageOffsetTable.
tile_start_offset[i] points to the start of the coded data of a tile. The position is given relative to the
referenced data in the ItemLocationBox associated to the image item. If a specific tile is empty and
does not contain image content, the tile is not coded and the tile_start_offset[i] entry shall be set
to 0xFFFFFFFF. This situation may occur when an image is generated on a canvas and certain portions
of the overall image only contain canvas with no image pixels. Readers shall interpret a tile_start_
offset[i] value equal to 0xFFFFFFFF as an empty tile with no media content.
tile_size[i] (if present) indicates the number of bytes of the coded tile bitstream.
The number of tile offsets stored in the table (NumTiles) is computed by

TileColumns = ceil(ImageSpatialExtentsProperty.image_width/
TiledImageConfigurationProperty.tile_width);
TileRows = ceil(ImageSpatialExtentsProperty.image_height/
TiledImageConfigurationProperty.tile_height);
NumTiles = TileColumns × TileRows
for (i=0; i< TiledImageConfigurationProperty.number_of_extra_dimensions; i++)
{
NumTiles = NumTiles × TiledImageConfigurationProperty.dimension_size[i];
}
TileColumns and TileRows are the number of tiles in a row within the overall image and the number of tiles
in a column within the overall image. image_width and image_height are the dimensions of the entire image

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
as specified in the associated ImageSpatialExtentsProperty item property. number_of_extra_dimensions
and dimension_size[] are defined in the TiledImageConfigurationProperty associated with the 'tili'
item. NumTiles represents the number of tiles in the entire tiled image item.
The entries in the offset table are ordered in row-major sequence. For a 2D image with a single coded layer,
they are indexed as [y][x], where:
x = tile column
y = tile row
For a 3D tiled image item, they are indexed as [z][y][x], where:
x = tile column
y = tile row
z = depth coordinate
For a general n-dimensional hyperrectangle, the tiles are indexed as [z ] [z ] .[z ] [z ][y][x], where z are
n-1 n-2 3 2 i
the n-2 extra dimensions.
x is the inner most looping variable, followed by y, and then z to z .
2 n-1
The coded tile data may be stored in the file in arbitrary order, resulting in the tile_start_offset entries
not necessarily being in increasing address order.
When size_field_length==0, the tile_size[i] variables are not present, and the decoder infers them from
the difference between the tile_start_offset entries. For the case where tiles are stored in sequential order
(flags & 0x10 == 0x10), the tile_size[i] is computed as tile_start_offset[i+1] - tile_start_offset[i],
except for the last tile, which extends until the end of the data. If the tiles are not stored in sequential order,
the decoder first sorts the tile_start_offset entries before computing the size from the offset differences.
In this case, the decoder cannot read the offset table on-demand. For on-demand applications, the tile sizes
should be included. When multiple tiles contain the same content, the tile_start_offset entries for these
tiles may point to the same data block. In this case, sequential ordering is not used.
NOTE In some cases, it is possible to encounter unspecified tile in a tiled image item. In such cases the rendering
of unspecified tiles is left to reader/player.

10.2.5.1
Add the following paragraph at the beginning of subclause 10.2.5.1:
This brand may be used when the item type of a coded image item allows multiple coded pictures within
the same coded image item, such as coded pictures representing different spatial scalability layers of the
same picture, but the coded image item actually contains only one coded picture that is intra coded. This
brand implies no semantics or reader requirements, when the item type of a coded image item does not
allow multiple coded pictures in the same coded image item.

Change the following sentence in subclause 10.2.5.1 from:
This brand enables file players to identify and decode HEIF files containing coded image items that only
contain one picture, and that picture is intra coded.
to:
This brand enables file players to identify and decode HEIF files containing coded image items that only
contain one coded picture and that coded picture is intra coded.

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
10.2.5.2
Change the following sentence in subclause 10.2.5.2 from:
A file having the '1pic' brand in the compatible_brands array of the FileTypeBox shall contain coded image
items that only contain one picture, and that picture is intra coded.
to:
A file having the '1pic' brand in the compatible_brands array of the FileTypeBox shall contain coded image
items that only contain one coded picture, and that coded picture is intra coded.

11.3.5
Add the following new subclauses after subclause 11.3.4:
11.3.5  Groups of regions
11.3.5.1  Union of regions entity group
A union of regions entity group ('unrg') indicates the union of all the regions represented by one or more
region items.
Each entity_id value in the entity group shall refer to a region item.
All the region items in the union of regions entity group shall be associated with the same image item, inside
which the regions are defined, using an item reference of type 'cdsc' from each region item to the same
image item.
If unique IDs are used:
— the union of regions entity group may also be associated with the image item inside which the regions
are defined using an item reference of type 'cdsc' from the union of regions entity group to the image
item,
— an annotation may be associated with the union of regions entity group by associating with it:
— an item property, using the ItemPropertyAssociationBox;
— a metadata item, using an item reference of type 'cdsc' from the metadata item to the union of
regions entity group; or
— an image item or another entity group, using an item reference of type 'eroi' from the union of
regions entity group to the image item or the other entity group.
11.3.5.2  Compound region entity group
A compound region entity group ('corg') associates one main region item with one or more region items.
It indicates an inclusion relationship between a main object covered by regions of a main entity and other
objects covered by regions described by one or more other entities, the main object logically including the
other objects.
NOTE 1 For example, a compound region entity group can be used to associate a main region corresponding to a
body with regions corresponding to body parts (e.g., the head, legs or arms of the body) to indicate that the body is
logically including the body parts.
The entities in a compound region entity group shall be region items.
The number of entities in a compound region entity group shall be at least 2. The first entity_id value shall
indicate the main region item. It indicates the region covering the main object that is logically including the
objects covered by the regions described by the second and following entity_ids.

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
This inclusion relationship does not convey information at the geometry level. A main region signaled as
including other regions by a compound region entity group may or may not geometrically include the other
regions.
NOTE 2 For example, the main region item corresponding to the first entity_id value can represent a bounding
box or a region encompassing partially the regions described by the region items corresponding to the second and
following entity_id values.
All the region items in the compound region entity group shall be associated with the same image item,
inside which the regions are defined, using an item reference of type 'cdsc' from each region item to the
same image item.
If unique IDs are used:
— the compound region entity group may also be associated with the image item, inside which the regions
are defined, using an item reference of type 'cdsc' from the compound region entity group to the image
item,
— an annotation may be associated with the compound region entity group by associating with it:
— an item property, using the ItemPropertyAssociationBox;
— a metadata item, using an item reference of type 'cdsc' from the metadata item to the compound
region entity group; or
— an image item or another entity group, using an item reference of type 'eroi' from the union of
regions entity group to the image item or the other entity group.

A.2.1
Add the following to the end of subcclause A.2.1:
When an ExifDataBlock is compressed using the deflate() algorithm defined in IETF RFC 1951, and the
resulting untimed compressed Exif metadata is stored as a metadata item, the item_type value shall be set
to 'dExf'.
J.2
Replace the contents of Clause J.2 with:
A file with a single coded image item, Exif metadata and T.35 metadata can be structured as follows:

FileTypeBox 'ftyp': major-brand='heic', compatible-brands='heic'
MetaBox 'meta': (container)
HandlerBox 'hdlr': 'pict'
PrimaryItemBox 'pitm': item_ID=1;
ItemInfoBox 'iinf': entry_count=3
1) 'infe': item_ID=1, item_type='hvc1';
2) 'infe': item_ID=2, item_type='Exif';
3) 'infe': item_ID=3, item_type='it35'; // ITU-T T.35

ItemLocationBox 'iloc': item_count=3
item_ID=1, extent_count=1, extent_offset=X, extent_length=Y;
item_ID=2, extent_count=1, extent_offset=P, extent_length=Q;
item_ID=3, extent_count=1, extent_offset=R, extent_length=S;

ItemReferenceBox 'iref':
referenceType='cdsc', from_item_ID=2, ref_count=1, to_item_ID=1;
referenceType='cdsc', from_item_ID=3, ref_count=1, to_item_ID=1;

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)

ItemPropertiesBox 'iprp':
ItemPropertyContainerBox 'ipco':
'hvcC'
'ispe'
ItemPropertyAssociation 'ipma': entry_count=1
1) item_ID=1, association_count=2
essential=1, property_index=1;
essential=0, property_index=2;

MediaDataBox 'mdat' or 'idat':
HEVC Image (at file offset X, with length Y)
Exif data block (at file offset P, with length Q)
T.35 data block (at file offset R, with length S)

J.8
Add a new Clause after Clause J.7:
J.8  Single image in a Low-overhead Image File
A file with a single VVC encoded image item can be structured as follows:

FileTypeBox 'ftyp':
major_brand='mif3'
minor_version='vvi3'
compatible_brands=''
MinimizedImageBox 'mini': (container)
version=0
explicit_codec_types_flag=0
float_flag=0
full_range_flag=1
alpha_flag=0
explicit_cicp_flag=0
hdr_flag=0
icc_flag=0
exif_flag=0
xmp_flag=0
chroma_subsampling=3
orientation_minus1=0
large_dimensions_flag=0
width_minus1=W
height_minus1=H
high_bit_depth_flag=0
large_codec_config_flag=0
large_item_data_flag=1
main_item_codec_config_size=C,
main_item_data_size_minus1=D,
main_item_codec_config=CompactVvcDecoderConfigurationRecord (with length C)
main_item_data=VVC Image Data (with length D+1)

© ISO/IEC 2026 – All rights reserved
ISO/IEC 23008-12:2025/Amd. 2:2026(en)
L.4.1
Rename the heading of subclause L.4.1 as:
L.4.1  'mif1'-compliant VVC image and image collection brands

L.4.3
Add the following new subclause after L.4.2.3:
L.4.3  'mif3'-compliant VVC image and image collection brand
L.4.3.1  Requirements on files
Files shall include 'mif3' as the major_brand and the brand 'vvi3' as the minor_version in the FileTypeBox
and conform to the specifications in O.2.1 of this document ('mif3' structural brand).
The 'vvi3' brand defines the coded image item type to be 'vvc1' as defined in L.2.2.1.2 and the alpha_item_
codec_config (if present), the gainmap_item_codec_config (if present) and the main_item_codec_config to
have the format defined by CompactVvcDecoderConfigurationRecord specified in L.4.3.3 of this document.
Files that include 'mif3' as the major_brand and the brand 'vvi3' as the minor_version in the FileTypeBox
shall comply with the specifications in L.2 and shall additionally be constrained as follows:
— explicit_codec_types_flag in the MinimizedImageBox shall be equal to 0,
— float_flag in the MinimizedImageBox shall be 0,
— If gainmap_item_data_size in the MinimizedImageBox is not 0, gainmap_float_flag in the
MinimizedImageBox shall be 0,
— If alpha_flag in the MinimizedImageBox is equal to 1, alpha_item_data_size in the MinimizedImageBox
shall not be equal to 0,
— The data of each codec configuration property shall use the syntax defined in subclause L.4.3.3.2 of this
document,
— Each coded image item shall contain a single IDR picture per layer,
— When multi_layer_flag is equal to 0, the data of each coded image item shall consist of only one NAL
unit excluding the length and NAL unit header fields, which are inferred by the reader as specified in
subclause L.4.3.2 of this document.,
— The equivalent file as specified in L.4.3.1 shall conform to the 'vvic' brand.
L.4.3.2  Requirements on readers
The requirements on readers specified in L.4.1.3 and O.2.1 of this document ('mif3' structural brand) shall
be supported.
As a response to 'mif3' as the major_brand and the brand 'vvi3' as the minor_version in the FileTypeBox,
the readers shall treat the file as if an equivalent file were created containing FileTypeBox with 'vvic' as the
equivalent major_brand and th
...