Advertisement

You're blocking ads, which pay for BlenderNation. Read about other ways to support us.

Blender 3.6 Alpha - Depth Map from 2D Images to 3D Objects

9

Gioxyer writes:

Depth Map are useful to use to give depth from 2D images with different formats from simple jpg to png high quality.

In this video you will see different steps that you can apply by starting by generating AI image or you can use personal ones. First of all you need to start by using AI Image Generator Tool called "Leonardo AI", you need to select a custom model in order to generate a good looking image.

After that you need to test one of these two tool: the first is called Midas and allow users to generate depth map by original image and the second is called Leia Pix Converter that can do also animation where you can set different parameters (speed, motion, blur).

I suggest also if you wanna do a good work to do Alpha channel with PNG (RGBA) with AI Image Eraser

Here there is the Youtube video that explain all the process to achieve a good 3D model from 2D image:

Let me know if you already test these technology and what do you think about the help of Artificial Intelligence to do 3D tasks.

About the Author

Italian UX/UI Designer and 3D Artist that believe new creators accelerate the world of tomorrow just now by making digital user experience that are beautiful design, simple to use and user friendly with a complete selection of new 3D Design trends, Web Design interfaces, UI Design Kit and Tutorial. We cannot predict how design will evolve, but that we can make it happen by giving a new shape to elements around us with our thoughts and our ideas that switch into projects that animate into our world life. Future is now, not tomorrow

9 Comments

  1. Art is not about results, but about process, because that's where growth lies...
    The result can be emulated by anyone and claimed to be their own, but only a real artist can experience the process.
    And that goes for all professions...

    Art is valued only because it takes Time, Patience and Practice to make something decent.
    When people without this dedication are also able to produce images just by typing "Keywords", art loses its value.

    If anyone thinks it's good the feeling of typing half a dozen words and in "half a minute" have something ready?
    You don't know how incredible is the "experience" of time lived when creating something real...
    The half minute used to create an Ai image is dead time...wasted...
    The Hours used to create a real Art, is experience... is spiritual and intellectual growth...

    No Ai, no machine can give someone the title of "artist", using images that not even Ai knows how to make, since they are stolen from real artists, obtained through texts and manipulated through filters..

    How could they put the name "Leonardo da Vince" in a program that doesn't make art, but creates images through the theft of other people's art...
    Leonardo da Vinte must now be brooding with shame and hatred in his coffin...

    I approve of Ai for helping artists, (UV, Retopology) but not for creating content using art of 2D artists without permission.

    People no longer have respect for what belongs to others...
    Forgive my vent for our brothers 2D artists

    • I agree in part that users can use to steal and use inappropriately, but it's like uploading on Internet something, you have your own rights (if set to Copyright material) and depends on people intelligence to take inspiration or use their own by breaking the rules.

      In general I can see people hate AI, but it happens also with the digital, when something new goes viral with a new market, in this case artist generator, but can be also code generator with developers (event of Chat GPT) tend to blame without even try that technology.

      I make another example when Google came out, more and more people at the beginning blame about it and now they use it, becouse became an habit. People when a new technlogy comes, they try to blame without even try and see the benefit.

      For example you can create your own branding by learning, asking question to Google, read books (physical or digital) and even asking to community what they think about a topic.

      It's the same thing with AI, you can take ispiration, upload your own 3D model and modify with AI (point generation maps) or also generate a spot, an ad with your own 2D graphic design and making catching advertisment just by clicking.

      This trend open new people to art, that can be bad, because people think that you can do with only a click, but if you know for example, a bit of photography you know the difference between a fixed lens or zoom lens. If you are curios enough, you can create even more opportunity.

      Probably AI can destroy markets, for artists, designers, architects, graphic designers, etc, but also create other more that are more competitive and wake up the market, artist need to adapt to new trends, not to be only more intelligent, but to adapt to any situation.

      I think people that blame can't solve problems, intelligence is also adapting, keep fightning and try to learn something positive from negativity from today's world.

      It's like the switch from analog to digital, why people love digital? Because you can do more with less energy, more efficient products, simple user interface and user experience, but probably the duration isn't the same of analog, but you've got more features with lower price in some cases.

      But this is my personal opinion, I think adapt to new change is a synonim of intelligence, keep researching something new, evolve and undestanding something that we know and apply our knowledge to make valuable products in the market.

  2. Who doesn't use Ai, Stay behind?
    Like this?

    I don't know any traditional artist, who was left behind when photography came along, or when digital art (Photoshop) came along...
    I know that traditional art was even more valued...

    Staying behind I think is who uses the AI a lot and with time loses its abilities...
    And with so much image generated by AI, it will devalue the market, and those with traditional skills will be well paid...

    I believe in it...
    I've seen this happen several times in history...
    The market will be saturated...
    Imagine an AI user generates 200 images a day, how will Virtual Disks look in a few years?
    With so much content of dubious quality on the internet, this will generate overload in many search systems and virtual disks...

    I believe this innovationwill blow up to nothing and everyone will return to good practices...

    Remember the NFT? Cryptocurrencies... ?
    Everything is moment...
    The only thing that isn't just timing is real talent and creativity.
    A real artist does not delegate the process to the machine, because growth resides in the process...
    Result is just consequence...

    I won't take any more of your time and mine...
    These conversations are unnecessary, they are just opinion...
    It's important that I respect yours and you respect mine.

    Thank you for your reply and healthy speech.
    I'm staying here.
    Luck on your way

    • I think for certain trends that expire like you mention NFT or Crypto, some days ago Instagram remove the possibility to sell NFT in less than an year, like 11 months ago now.

      But other can be effective to learn something new, for example Deepl in 2017-2018 generate more competitive and intelligent, but in some cases they wrong the translate, human in certain categories isn't comparable, for complex things like generate a storytelling and have emotions.

      For example, the other day, I tried on Chat Gpt and ask a informative task and it wrong completly the answer and even if you correct him, it will continue do the same error.

      On storytelling for example it's better human, because can feel in certain situation different emotions, you can also undestand the scenario, etc.

      But for what I see in some cases AI is implemented in software like Photoshop, etc, but for what I see people think that with post-production you can achieve a good photos, but rules are so important and they generate goes without knowing the basic principles.

      I think even when you do a research, you need to select information, but as you mention before when people upload more with AI, it will become more and more frustrating for the user to select good sources.

      There are some AI tools that scan the theme of the website, but sometimes they invent something that there isn't in the text and can be dangerous in long run.

      So basically anyone can write an article without even know the topic and rank on Google, because with other AI tools you can reduce AI Recognition and cheat with an human writing, but you can notice that AI generally often invents things.

      In this case when users has so many information around their life, they will make confusion.

      There are pro and cons for AI, the negative side are so dangerous for artists and the rules for any countries keep changing, but not as fast as digital markets.

      I agree the fact when you create something yours, unique, give 10 times more than an AI generated, for example modelling a car, sculpting a character, meeting new artists, designers, architect can't be replaced by AI.

      Art and emotions are extremly connected togheter and when you know how you do something, you can explain better what you've done, you can be inspired by other artists and you know all the sources to take information from.

      It is a powerful tool, but at the same time if it will be used with a destructive goal, artists probably will switch interest in something that maybe like less and less.

      Thanks for your opinion,
      Gioxyer

  3. Casey R Williams on

    AI is to visual art what sampling was to music. Some see it as stealing, others see it as creating new art. The only thing that is a shame to me is that the artists' names were attached to the work the models were trained on.

    • There are some rules that try to regulate copyright issues, but digital market evolve fast, so it can becomes difficult to safeguard the heritage of masterpieces of artist.

      In general for what I see it can be used in other direction like as inspiration model to generate new ideas, the problem is that you didn't know the sources and the reference, because maybe the AI combine other style.

  4. Oliver H Bailey on

    As both a scientist and artist, I find extreme pleasure and value in only using my own works from start to finish. So long as no digital copies get into the public domain, duplicating the exact processes is impossible to obtain even a close copy. a true artist will take care in their work to perfection. i have pieces that were started 12-15 years ago that are just reaching a point of artistic worthiness. AI wil take two minutes to produce something an average user will have a 15 minute attention span to enjoy. They will become bored for the lack of the very edperience in creating the work is missing. I neve see the quality where there was no thought contributed to obtaing the experience.

    • I think in some cases can be helpful using AI to take inspiration, for the process can be confusing, especially for beginners that need to work more to undestand basic principles and multiple times AI can't be predicted and also when it makes mistakes, it keeps going with that, without recognise the main error.

      Considering that at initial state, I think will become more relevant in next years, everything we have today use AI from smartphones to cameras with eye recognition, in some cases having too much information can cause the opposite thing, so artists, developers and all around these topic will become more addicted to the technology.

      So basically people will think less and less without even know what they have done with AI, but if you have enough experience you can use as your own as an advantage for inspiration, for the complete creation I think developing our own personal skills can't be replaced.

      A good thing that humans have is emotions that depends on output that we receive and also the storytelling behing a work, a project or a presentation, in that sector AI generally invent something that isn't the same level and doesn't have empathy to the person who listen or read your creation.

      Thanks for your point of view

  5. At the time of this writing, based on 4 images that I tried with each service, Leiapix has given me great depth maps that I've used in Blender ($7.99 for 10 exports). Midas v3.1 has produced poor results for the same few images, but it's free.

Leave A Reply

To add a profile picture to your message, register your email address with Gravatar.com. To protect your email address, create an account on BlenderNation and log in when posting a message.

Advertisement

×