Home – New Forums Tech talk Change colours on live TV

  • This topic is empty.
Viewing 3 posts - 16 through 18 (of 18 total)
  • Author
    Posts
  • #1175929
    quickbas
    Member
    • Total posts: 53
    Up
    0
    ::

    Still a mystery to me.

    Do you want to change the colours of the entire TV screen or just a particular part of the screen (eg an overlay)?

    You want to do this on a user-by-user basis? Or show-by-show? Or with a remote when watcher chooses?

    Modern TVs have many options to change the colour map (brighter/darker, cooler/warmer, vibrant/desaturated). Sometime via remote. But not unless the user chooses to do so.

    The only way I could see this being automated (if that is what you are after) is some sort of modulator between the set-top-box and the screen.

    #1175930
    ozcos
    Member
    • Total posts: 32
    Up
    0
    ::
    The Flagman, post: 204751 wrote:
    Hi Fellow Soloists

    Is it possible to change the colour of an object whilst watching live TV ?? A challenge no doubt

    The Flagman

    Hi Flagman
    To answer your question directly I will say yes

    The 2 questions that will need to be answered is
    1. Budget – what is your proposed budget or do you require funding?
    2. How is the desired outcome achieved?

    Unfortunately so often with new ideas both above questions are directly related.
    You may not now the cost of doing / producing something until you know how to achieve the desired outcome.

    This is where prototyping comes in – if you require funding is will be much easier to obtain with a “working prototype” even if it is chunky and not 100% – just something that can prove that “it can be done with some tweeking to others that don’t understand (or need to understand) the technology behind the system”

    That being said, it is obvious that I would need to ask the question – what are the details for your idea and how do see it working / used?

    Prob something you will not answer here. (and understood)

    Bit of details of my thinking,

    Yes the concepts known in video production will come into play but only at a top level “the action”

    And “it’s really hard for computer vision to pick out discreet objects and know what they are.” is also very true.

    These are global responses based on the limited information provided.

    It is also very true regarding the statement: “If an environment can be controlled”.
    Next thing to understand when working on this solution is to factor in or out what can be controlled (this will provide the “how to’s)

    The most significate factor in the controlling the environment is knowing exactly how the system will be implement (the full details).

    The simple way to explain:
    consider the idea of engineering a bridge over a river, if you ask 100 engineers to submit a design then you will receive 100 different designs which 90 of them will not be suitable for your outcome.

    If you ask 100 engineering to submit a design for a bridge for only pedestrians and bicycles then you will receive 100 suitable designs (you just partially controlled the environment) but 90 of them won’t suitable for the environment you can’t control, eg the width of the river, council laws regarding cosmetics etc.

    If you ask 100 engineers to submit a design for a bridge for pedestrians and bikes, must have a roof for rainy weather, must be of a suspension design that does not obstruct river traffic, no lower than 50m from the average yearly high tide point, made from steel, within the local council cosmetic design restraints and so on, chances are you will receive 100 designs that are all suitable.

    Each time you provide your requirements you are controlling the environment simplifying the design and increasing the probability of achieving the desired outcome within the project restraints (budget / time). The result will be making a decision based only on budget and time (not how it can be done)

    Your object detection system (which is what you are looking for – not a colour changing system) system will work very much the same way, until you know and understand the environment it will be difficult to design a suitable system (or even know where to start).

    There are many ways to “detect” an object and make decisions within such a proposed system, if you research down the lines of video surveillance systems including open source solutions in this area (which are using live feeds) and visual AI concepts plus a level of probability (prediction) then the possibility of achieving the desired outcome increases.

    If you don’t have the skills or knowledge to design such a system then you will need to employ, partner, or at least talk to somebody else at some point.

    I have a very genuine interest in your idea at many levels (including business), and do provide an invite only mechanism for technology design and engineering discussion at the design level specifically for new ideas and under confidential contracts this includes conceptual designs right through to funding and prototyping.

    Feel free to PM me with your contact details and I can provide you further information.
    Cheers,

    #1175931
    The Flagman
    Member
    • Total posts: 9
    Up
    0
    ::

    Good morning

    Thankyou for your reply Please email me at [email protected] and I will provide more info

    Cheers allan

Viewing 3 posts - 16 through 18 (of 18 total)
  • You must be logged in to reply to this topic.