Commit 2f177093 authored by Abhishek Dutta's avatar Abhishek Dutta
Browse files

updated readme and associated files

parent 6e802ab4
Each version of VIA maintains their own copy of CHANGELOG in the following files:
* [via-1.x.y/CHANGELOG.md](via-1.x.y/CHANGELOG.md)
* [via-2.x.y/CHANGELOG.md](via-2.x.y/CHANGELOG.md)
* [via-3.x.y/CHANGELOG.md](via-3.x.y/CHANGELOG.md)
## [3.0.0] - 17 May 2019
* development of via-3.x.y began in Dec. 2018 with the aim to create a basic audio and video annotator
* no external dependency, self contained and easy to use
* light weight (complete application fits in a single html file of size ~ 275 KB)
* manual annotation of local (using file:// uri) and remote (using http:// uri) files
* supports definition of both temporal segmentation (e.g. from 3.2sec to 8.7sec) and spatial region (e.g. bounding box in still video frame)
* keyboard shorcuts allow human annotators to be quick with manual annotation
* annotations can be exported in CSV format
* to reduce cognitive load on human annotators, VIA3 support for attributes of different types (plain text, drop down, radio, checkbox)
* self contained demo for audio and video annotation
## [2.0.7] - May 01, 2019
- regions cannot be moved beyond image boundary (issue 173, thank you @zenggyu for bug report)
- user's mouse cursor coordinate visible (feature request 172)
- shortcut 'd' now works even when multiple regions are selected (merge request 15)
## [2.0.6] - March 25, 2019
* fixed wrong region location for high zoom levels for large images (Closes: #166, thank you @fredowski)
* Update region colour based on value of a region attribute (issue 175)
* Relative paths now gets resolved (issue 171)
* Grid view filter issue with "only first,middle and last image" fixed (issue 167)
* grid view should only consider files filtered using filename regex in sidebar (issue 176)
## [2.0.5] - December 03, 2018
* reverted back to a.dispatchEvent() from a.link() which does not work in Firefox 63
## [2.0.4] - November 29, 2018
* fixed annotation download issue encountered in Safari browser (issue 162, thank you @fredowski)
* region selection state flag is now cleared and re-initialized when a new image is loaded (issue 159)
* fixed missing state variable `_via_is_ctrl_pressed` issue (issue 160)
* when drawing polyshape, [Enter] finishes drawing and [Backspace] deletes last vertex (issue 163, 164)
* improved subselection of images in a group (image grid view)
* keyboard handling for zoom +/-/= changed from e.which() to e.key() (merge request 10 contributed by @fredowski)
* simplified and improved keyboard shortcut keys, global shortcut keys (like next/prev image) are active even when canvas is not in focus
* fix issues with on-image annotation editor (issue 155)
* on-image annotation editor shown automatically when a new region is drawn
* fixed issue with attributes of type checkbox (issue 158)
* on image annotation editor is not shown if region attributes are not defined
## [2.0.3] - October 25, 2018 (REVOKED: Bug with region resize)
* REVOKED
## [2.0.2] - October 10, 2018
* bounding boxes can now be drawn using touchscreen in mobile devices (more tests needed)
* annotation editor now supports single region on-image editor (contributed by GYOUNG-YOON RYOO)
* anotations downloaded as image is now unaffected by current zoom level(issue 154)
## [2.0.1] - September 26, 2018
* clicking or selecting a row in annotation editor (at bottom) should select the corresponding region. (issue 137)
* converted svg logo to use standard font
* zoom is enabled with [Ctrl] + mouse wheel (issue 143)
* fixed polygon first vertex resize issue (issue 146)
* annotations overlaid on image can now be downloaded as an image (issue 24); does not work with images added as URL (browser security)
* import annotations: handle entries for which file_size = -1 (i.e. unknown)
* When zoom is active, Shift + mouse wheel moves horizontal scroll bar
## [2.0.0] - July 15, 2018
* rectangles can now be resized from edges (issue 120)
* added POLYLINE shape
- polygon region shape coordinates set does not end with the first coordinate (closed loop is defined by the shape: polygon, polyline)
* renamed many methods and variables
* attributes editor
- added _via_attributes{'region','file'} and removed _via_region_attributes, _via_file_attributes
- attributes add/remove/update from left sidebar panel
- attribute types: text (default), checkbox, radio, dropdown, image (from url or base64 encoded image data)
* annotation editor
- similar to previous versions (spreadsheet like editor) supporting attribute types: text, checkbox, radio, etc
- tabbed panel for region and file attribute editor
- change size of contents and panel
* via project save/load
- everything (annotations, attributes, application settings, etc) can now be saved as a VIA project (a json file)
* exported csv annotation header line does not contain "#" (this may be a breaking change) (issue 122)
* zoom level does not reset when navigating prev/next images (issue 125)
* VIA project: open/save
- image file list can be filtered using regular expression
- preset filters for image file list
- shortcuts to add files to project (local, url) and remove files from project
- files in local disk drive can also be added using absolute path (e.g. /home/tlm/image1.jpg)
- project settings
* paste regions in multiple images and option to undo this operation (issue 123)
* (simplified code) image content is shown using <img> instead of <canvas> (region shapes still shown using <canvas>)
* image grid to perform bulk operations on a set of images (like changing file attribute of multiple images)
- group by operation supported on file and region attributes
- multiple groups can be created
- next/prev page navigation
- increase/decrease size of image shown in image grid
* images are preloaded and buffered to allow faster and smoother prev/next image navigation
* image data can be pulled from custom locations in local disk in order to resolve image files
* shortcut toolbar (in top) now uses icons from material icon set (https://material.io/icons/)
* updated ImageMetadata() and RegionMetadata()
* removed persistence of annotation data in browser cache (i.e. localStorage)
- this feature proved difficult to maintain for cross browser support
- uses "beforeunload" event to warn users of data loss
* minor improvements
- updated region shape color to yellow
- selected regions can be precisely moved by 1 pixel using arrow keys (press Shift to move by 10 pixel)
- move multiple selected regions
- {polygon, polyline, point} regions can now be drawn inside existing regions (i.e. nested regions)
- both polygon and polyline shape drawing operation is now ended by pressing [Enter] key
- when region is selected, mouse wheel can be used to change the region label
- center of circle and ellipse shapes remain fixed to mousedown location (major issue in 1.0.x versions)
* updated regression test cases
* polygon vertex can be added and removed (issue 134)
## [1.0.6] - June 15, 2018
* a patch from Stefan Mihaila which requires polygon shape to have at least 3 points.
* rectangles can now be resized from edges
* added POLYLINE shape
* image file list can be filtered using regular expression
* renamed methods and variable
- _via_reload_img_table : _via_reload_img_fn_list_table
- reload_img_table() : reload_img_fn_list_table()
- _via_loaded_img_table_html : _via_loaded_img_fn_list_table_html
## [1.0.5] - January 16, 2017
* (code contributions from Stefan Mihaila) via.js codebase improvement, wider web browser support (IE 10, IE 11 and Opera 12)
* added Contributors.md file to record contributions to VIA codebase
* removed 'localStorage.clear()' to avoid SecurityError in Safari browser (issue 85 and 108)
## [1.0.4] - October 17, 2017
* fixed polygon copy/paste/resize issue (issue 107)
## [1.0.3] - August 07, 2017
* CSV export now does not add extra comma to each line (issue 103)
## [1.0.2] - August 04, 2017
* removed free resize of ellipse from any edge (issue 100)
* fixed free resize of rectangle (issue 101)
* fixed 1-pixel bug (first set image space coordinate, then set canvas coordinate. see issue 96) for region resize and move
* press Ctrl while resizing to preserve the aspect ratio of rectangle (issue 98)
* fixed issue with CSV files containing newline character \r or \r\n (issue 102)
* top menu bar remains consistent event when the user scrolls the window
## [1.0.1] - June 11, 2017
* fixed issue 33 : Annotations cannot be imported from file of type application/vnd.ms-excel
* fixed issue 96 : A major bug in how canvas coordinates are computed
## [1.0.0] - April 04, 2017
* file-attributes support added (useful for weakly supervised learning)
* spreadsheet like editor for region and file attributes
* visualization of loaded image list improved
* user annotation data cached in browser's localStorage (for data recovery on browser crash)
* zoom in/out support
* improved performance using multi-layered canvas for image and annotations
* new user interface layout (added toolbar on top navigation panel)
* added Getting Started guide and License to help menu
* CSV import/export now conforms to RFC 4180 standard
* added some basic unit tests
* added support for point regions (useful for landmark annotations)
## [1.0.0-beta] - 2017-03-15
* beta release for VIA 1.0.0
## [0.1b] - 2016-10-24
* first release of VGG image annotator
* supports following region shape: rectangle, circle, ellipse, polygon
* contains basic image region operations such as move, resize, delete
* Ctrl a/c/v to select all, copy and paste image regions
* import/export of region data from/to text file in csv,json format
* display list of loaded images
VIA software is developed using HTML, CSS and Javascript and is based solely on
standard features available in modern web browsers.
VIA does not depend on any external libraries. These design
decisions has helped us create a very light weight and feature rich manaul
annotation software that can run on most modern web browsers without
requiring any installation or setup. The full VIA software sprouted from an
[early prototype](http://www.robots.ox.ac.uk/~vgg/software/via/via-0.0.1.txt)
of VIA which implemented a minimal -- yet functional -- image annotation tool
using only 40 lines of HTML/CSS/Javascript code that runs as
an offline application in most modern web browsers. This early prototype
provides a springboard for understanding the current codebase of VIA which
is just an extension of the early prototype. The [introductory tutorials](https://developer.mozilla.org/en-US/docs/Web/HTML)
prepared by Mozilla is also very helpful in understanding the basic concepts of
HTML/CSS/Javascript platform.
The development of VIA software began in August 2016 and the first public
release of Version~1 was made in April 2017. Many new advanced features
for image annotation were introduced in Version~2 which was released in
June 2018. The recently released Version~3 supports annotation of audio and video.
Therefore, the VIA source code repository contains a separate folder for each
major version of VIA: [via-1.x.y](via-1.x.y/), [via-2.x.y](via-2.x.y/) and [via-3.x.y](via-3.x.y/).
The development of each version is carried out in a separate branch (e.g. [via-2.x.y branch](https://gitlab.com/vgg/via/tree/via-2.x.y).
If you wish to contribute code to VIA (we encourage you to do so), please send
a pull request to one of the branches. Please do not send pull requests to the
[master branch](https://gitlab.com/vgg/via/tree/master).
We have prepared the following code documentation for each major version of VIA:
* [Code Documentation for via-1.x.y](https://gitlab.com/vgg/via/blob/master/via-1.x.y/CodeDoc.md)
* [Code Documentation for via-2.x.y](https://gitlab.com/vgg/via/blob/master/via-2.x.y/CodeDoc.md)
* [Code Documentation for via-3.x.y](https://gitlab.com/vgg/via/blob/master/via-3.x.y/CodeDoc.md)
Software bug reports and feature requests should be
[submitted here](https://gitlab.com/vgg/via/issues/new) (requires a gitlab account).
```
Author: Abhishek Dutta
Version: 20 May 2019
```
......@@ -3,8 +3,13 @@ VIA is an open source project and we welcome all forms of contributions
(code update, documentation, bug report, etc.) from users. All the contributions
must adhere to the existing [license](LICENSE) of the VIA project.
Each version of the VIA software maintains its own list of contributors in the
The development of VIA software began in August 2016 and the first public
release of Version~1 was made in April 2017. Many new advanced features
for image annotation were introduced in Version~2 which was released in
June 2018. The recently released Version~3 supports annotation of audio and video.
Each version of the VIA software maintains a list of their contributors in the
following files:
* [via-1.x.y/Contributors.md](via-1.x.y/Contributors.md)
* [via-2.x.y/Contributors.md](via-2.x.y/Contributors.md)
* [via-3.x.y/Contributors.md](via-3.x.y/Contributors.md)
......
......@@ -55,6 +55,40 @@ We welcome all forms of contributions (code update, documentation, bug reports,
Such contributions must must adhere to the existing [license](https://gitlab.com/vgg/via/blob/master/LICENSE) of
the VIA project.
## Developer's Resources
VIA software is developed using HTML, CSS and Javascript and is based solely on
standard features available in modern web browsers.
VIA does not depend on any external libraries. These design
decisions has helped us create a very light weight and feature rich manaul
annotation software that can run on most modern web browsers without
requiring any installation or setup. The full VIA software sprouted from an
[early prototype](http://www.robots.ox.ac.uk/~vgg/software/via/via-0.0.1.txt)
of VIA which implemented a minimal -- yet functional -- image annotation tool
using only 40 lines of HTML/CSS/Javascript code that runs as
an offline application in most modern web browsers. This early prototype
provides a springboard for understanding the current codebase of VIA which
is just an extension of the early prototype. The [introductory tutorials](https://developer.mozilla.org/en-US/docs/Web/HTML)
prepared by Mozilla is also very helpful in understanding the basic concepts of
HTML/CSS/Javascript platform.
The VIA source code repository contains a separate folder for each
major version of VIA: [via-1.x.y](via-1.x.y/), [via-2.x.y](via-2.x.y/) and [via-3.x.y](via-3.x.y/).
The development of each version is carried out in a separate branch (e.g. [via-2.x.y branch](https://gitlab.com/vgg/via/tree/via-2.x.y).
If you wish to contribute code to VIA (we encourage you to do so), please send
a pull request to one of the branches. Please do not send pull requests to the
[master branch](https://gitlab.com/vgg/via/tree/master). All contributions
must adhere to the existing [license](LICENSE) of the VIA project.
We have prepared the following code documentation for each major version of VIA:
* [Code Documentation for via-1.x.y](https://gitlab.com/vgg/via/blob/master/via-1.x.y/CodeDoc.md)
* [Code Documentation for via-2.x.y](https://gitlab.com/vgg/via/blob/master/via-2.x.y/CodeDoc.md)
* [Code Documentation for via-3.x.y](https://gitlab.com/vgg/via/blob/master/via-3.x.y/CodeDoc.md)
[Code review of via-1.0.0](https://gitlab.com/vgg/via/issues/54) was compelted
by @ecoto in Feb. 2017. We are [always looking](https://gitlab.com/vgg/via/issues/152)
for external developers who can help us with regular code reviews of the VIA software
code.
## Citation
If you use this software, please cite it as shown below and acknowledge the Seebibyte grant as follows: "Development and maintenance of VGG Image Annotator (VIA) is supported by EPSRC programme grant Seebibyte: Visual Search for the Era of Big Data (EP/M013774/1)"
```
......@@ -74,9 +108,6 @@ If you use this software, please cite it as shown below and acknowledge the Seeb
}
```
## Developer Resources
**Please send all pull requests for a specific version (e.g. via-2.x.y) to their respective branch (e.g. branch via-2.x.y). Please **do not** send pull requests to the master branch. All contributions made to VIA code repository will be licensed under the [BSD-2 clause license](https://gitlab.com/vgg/via/blob/master/LICENSE).**
## Contact
Contact [Abhishek Dutta](adutta_remove_me_@robots.ox.ac.uk) for any queries or feedback related to this application.
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment