"Grainy" movies

[seplute]

Opening a new thread to separate “grainy” movies from “blurry” movies

@Plenum

Hi, All -I was originally complaining(?) about “grainy” videos - and it appears that “blurry” videos are entering the topic. But, no, the two are distinct, to me. I’m compiling a list of the “worst grainy videos” (in my opinion) as I work through, so there’ll be another document (maybe in 7-10 days) to go with the already emailed to sepulte about three days ago.–As mentioned in the email, grainy videos, to me, do no good when it comes to analyzing excellent, detailed videos […]
I would save the grainy videos for another, say, introductory category of analysis.

((Blurry movies is another topic.))

Happy Holidays!

Plenum

@gcalkins

These may be examples of what @Plenum is seeing from the High Fat Dataset. We had more of these types of movies earlier this year. Will continue to flag them when we see them.

I have seen similar movies that had this same feature along the bottom edge. No defined vessel to annotate.

@Plenum, @gcalkins and all – let’s continue this topic here.

@Plenum – we will most likely get back to you after the holidays, but yes please do send more examples if you have them!

[seplute]

Hello!! I am finally here with some answers…

Thank you for sending all the examples @Plenum Here’s what @pietro said about the importance/inevitability of grainy movies:

The lack of consistent quality in the movies with respect to blurriness, graininess, overexposure, etc. can be frustrating, and over time, we are improving our methods to get higher quality data. But having a variety of calibration videos that includes sub-optimal image quality is needed to help train people so they can provide more accurate answers when confronted with real vessel images that have similar characteristics. We are doing our best to “meet in the middle” - that is, improve image quality, and train users to annotate when image quality is sub-optimal.

Basically, in some cases we simply have to deal with the data as it comes from the lab, and sometimes the imaging and/or preprocessing conditions lead to worse quality movies. But as long as we can tell flowing from stalled in them, we’re good to go, and its invaluable for our purpose!!

Thank you for bearing with us! Real is not always beautiful, haha!

[Plenum]

Hi Seplute, and thanks for writing the note, Pietro…

Well, after a nice break over holidays and thinking about all this, I understand that it’s all new for many of us and that we have to work with what we’ve got - sort of roll with the punches and the limits at hand. Frankly, I’m really impressed with some of the images, the clarity of them.

Some of my comments probably stemmed from some frustration at the differences between them and what were still-developing techniques to obtain images of particular vessels and capillary beds.

But, hey, I feel fine with my progress, and hope to continue well into the future. It’s all very interesting!

Yours, Tom F.

[seplute]

Happy to hear that @Plenum, and thank you for being part of the crowd

[chairgaf]

Hello folks,

I now understand completely, what Seplute is saying. We (catchers) need the poorer quality movies as ways for us to increase our scrutiny powers of ANY kind or movie. Good OR bad quality.

This also helps the admins to see just bad a movie can be, before it confounds us catchers completely.

I’ve also gotten frustrated over these poor quality movies. But as said above, I now see that, even though the Team may be working on eliminating most of the bad quality movies, we still need some to help keep us catchers sharp eyed.
Cheers,
chairgaf