Thank's for the great project, I think it's really amazing!

I'm exploring a dataset for which the time-view is not very interesting, and I would like to start off on the tsne-view instead. I was wondering, is there an easy way to accomplish this?

This is rather a feature request than a bug, but the preview feature in the normal filter is quite nice, as you can quickly go hover through the filters. This is not working on the hierarchical filters.

Hey Christopher,

this is a great project! We modified the Vikus Viewer and used it as a part of our project.

Some minor things:

• It would be great to remove the console.log statements or introduce a debug config flag. (Maybe you can have a dev branch where you keep them).
• Also, I am not sure if it is necessary to have the time information from the tsne
• In js/canvas.js we had some errors with selectedImage. In line 374 and 663.

I removed the comments and fixed the selectedImage in this repo. If you agree, with the removal of the comments, I can a make Pull request :).

I'd love to be able to see multiple timeline views (visualize a collection by creation date and also by accession date, for instance). Would be great to add new tabs on the left (along with "time" and "similarity") for additional timelines.

This would also allow for a grid view (a timeline view with a single column) that visualizes a collection by color, or brightness, etc. (Sort the data.csv by hue and present as a single-column timeline.)

Thanks!

My Lego Sets gallery is no longer online, so please remove it from the readme.

Hello!
The latest update that changes "tsne" to "similarity" in the .navi sidebar (2e5879c) in index.html appears to break the TSNE visualization unless viz.js or canvas.js are modified, since state.mode is set using canvas.setMode()

with the text of the label...

...and then referenced directly to determine whether to project TSNE...

Would using the button's class (.button.tsne, .button.time) instead of the text make more sense? Or is it better to just change the if statement on line 660 of canvas.js?
(Or am I missing something else here? Figuring out Vikus for the first time!)

First of all thank you for this great viewer, I'm really loving the potential of this!

I'm trying it out with a dataset of around 11000 images. And it all works really smooth after the first page load.
However, performance drops once I clicked one of the images to see the detail view and zoom out again. (latest chrome on mac 10.14)
Please see my example here: http://nijdam.nu/vikus-viewer/

I did checkout out the tsne branch to see if that might improve things, but that made no difference.

On a side note, do you have a roadmap? because I'm actually wanting to sort on two columns like you do on https://uclab.fh-potsdam.de/coins/ and on other examples you also have a tsne and grid layout I'm wanting to incorporate. (give someone a finger and they want the whole hand :-)

Also I commented out the if statement on line 470 in canvas.js because otherwise I did not have any data in the detail page. (now I'm writing this I realize it's because my columns in data.csv are not prefixed with a _)

for (field in selectedImage) {
    //if (field[0] === '_')
        detailData[field] = selectedImage[field]
}

I understand there is still some work to do, and I know that you do this on a voluntairy basis so thank you for this! Really looking forward to future releases.

EDIT:
It has something to do with the line I changed to make the detail view working, really weird. Performance is ok when I revert canvas.js but now I have no data in the #detail view.

EDIT2:

I did not notice the columns year, id and keywords are automatically available as _keywords etc. I thought since I have a column keywords in data.csv the source should also be keywords, but I was wrong, source should be _keywords.

Now it looks like it is working fine, including zoom in and out.

So the only question I have left is when the other layout options will be available :-)

Sincerely,

How much work, do you think, it would be add to the Vikus viewer a second and identical keywords bar that has the operator NOT built in. So, hitting a keyword there will exclude all items having the keyword -- instead of hiding all items that don't have the respective keyword (as it works now). The two bars don't necessarily need to work in combination but one could just add a toggle for the operator "NOT". And the same would also work for "OR". This would make visual exploration much more powerful.

When zooming out fast, while filtering is active, some of the images in the lower area are not in filter-out-state

When you zoom on portrait-mode images, they are too big for the canvas, when you zoom out, the info-box disappears, this means you cannot see the full image while also seeing the info-box.

When the hierarchical Filters are in their unfiltered state, the font-size differs, i guess the font-size depends on the largest internal group.

But as soon as one selects any top-level filter, all top-levels become the same font-size.

I think its a bit confusing that the top-levels have different font-sizes, and that those sizes change once something is selected. But maybe this is personal taste. Maybe there could be a config option to toggle this?!

Hello @cpietsch, first of all thank you for this beautiful piece of code you put together, it's being really helpful for more than one project I'm working on.

I'm trying to get the map branch to work with my data and everything looks fine except for one or two issues: the sprites don't seem to be loaded either on the timeline or the map, even though they work fine with the "default" vikus-viewer. The console is not giving me any errors about that. Other than that, when I click on a sprite placeholder on the map, it zooms in way more than expected, rendering bigImage larger than it's supposed to (that is, fitting the window left of sidebar detail). Here are some screenshots:

Do you know what might be wrong with my instance? I tried messing with the hardcoded scales and ratios such as zoomedToImageScale and zoomToImage related vars but couldn't really get it to behave the way I'd like.

Thanks again!

In tsne view, I'd like to be able to filter the view, zoom in, find an object, then press "x" to clear all filters, and see the object surrounded by all similar objects. (Currently you need to find the object, then zoom out, cancel the filters, then zoom back in...)

Alternatively, it would be great to access the filters somehow while zoomed in.

Thanks!

timeline.csv links to a non-existing timeline.json

One way to facilitate sharing might be to expose selected tags in the URL. Basically you need to a) look in the querystring for tags and add those to the array of selected tags and b) update the querystring when selecting/deselecting tags.

I ended up in another dead end: how could I display multiple pages of the same document (same data.csv line)? I'm trying to obtain the same navigation as in the https://vikusviewer.fh-potsdam.de/recs/vis/ (Seite with multiple pages and the ability to click trough previous and next page). In my data.csv I have an id column followed by several columns without any other header, but with id names in them (like in the image bellow):

Any advice would be really helpful.

Especially when having multiple hierarchical dimensions....

keyboard navigation with plus, minus and arrow keys; maybe tab for next photo

• I saw this as a project card. Any way I could give a hand in helping with this?

When one clicks on a multipage - next image button, the first image is visible in low-res, while the other image loads...

When I launch this I am unable to do anything and the page is black/says Powered by VIKUS Viewer. How do I actually do things with this? Am I doing something wrong?

I have prepared the sprites with vikus-viewer-script. So there's a spritesheet.json and sprites-1024-0.jpg upto sprites-1024-60.jpg.

But when I load https://lab.coret.org/vikus/ (with config.json) I do not see these sprites. If I zoom in enough, the images from the 1024 folder are shown (same for the map view). When I click a image, the right metadata is shown in the right panel.

What am I doing wrong with the sprites?