ฝัง
- เผยแพร่เมื่อ 19 พ.ค. 2024
- Check out VIDU here:
www.shengshu-ai.com/home?ref=...
📧Join my newsletter
delightfuldesign.eo.page/w7tf5
---
👨🏫Check out my AI courses:
www.udemy.com/user/samson-vow...
---
💼Business inquiries
samsonvowles@gmail.com
---
--- My top resources:
📖 Grab My AI Secrets! Dive into my handpicked prompts e-book.
delightfuldesign.eo.page/6sw6r
🎨 Unleash Your Creativity! Experience Adobe’s cutting-edge AI tools.
tinyurl.com/2sj2kwk6
🤖 Boost Your Efficiency! Automate with ClickUp.
tinyurl.com/4p9yxew5
🎤 Find Your AI Voice! Clone your voice with precision.
elevenlabs.io/?via=b38559
🎥 Revamp Your Content! Edit shorts effortlessly with AI.
www.opus.pro/?via=7fcbf8
✍️ Supercharge Your Writing! Generate content the smart way.
www.jasper.ai/?fpr=samson
📹 AI-Powered Videos! Elevate your content with Kaiber.
kaiber.ai/?via=samson
📢 Ads Made Easy! Create compelling ads in a flash.
free-trial.adcreative.ai/2yui...
👤 Stand Out Online! Craft your unique AI avatar.
www.synthesia.io/?via=delight...
🎬 Level Up Your Channel! Get coached by AI on TH-cam growth.
vidiq.com/samsonvowles
🖥️ Present Like a Pro! AI-driven slides for impactful presentations.
slidesai.io/?via=samuel-vowles - วิทยาศาสตร์และเทคโนโลยี
The race in AI-generated videos is truly accelerating the technology. Comparing the quality of AI videos from a year ago to now, I've seen a dramatic shift in the way I perceive and create art. As a 3D VFX artist with a strong background in cinematography and photography, this past year has been transformative. VIDU looks promising, and I'm eager to see its impact when released. If it launches before SORA, it could certainly push OpenAI to respond, possibly with a lighter version of SORA. They might even offer different versions at varying price points, tailored to different users. The most advanced version might be priced for big production studios. It's hard to imagine the tools we'll have in a year. I believe achieving 100% realism will soon become the norm.
“You can't look at the competition and say you're going to do it better. You have to look at the competition and say you're going to do it differently” - Steve Jobs
So true!
These AI videos are getting absolutely insane. Thanks for the update, and GREAT presentation.
Thanks for your support!
Awesome vid. SORA should hurry because most sci-fi films don't use a lot of long scenes, just a dozen short scenes strung together with VIDU could literally compete with Hollywood's finest. And I can't wait.
What you mean, I think, is "shots". not "scenes"> Movies nowadays don't have long shots, they're often short. less than 10 seconds, unless it's extended dialogue. Scenes are composed of shots and movies are composed of scenes that contain multiples shots.
@@MrForestExplorer Strung together into a scene. Several shots can become a scene. I've done with stock footage. Say you generate the first 15 seconds, then edit focus to another none AI shot, another 15 sec AI shot, repeat a dozen times. All put together looks like a long scene that looks like a Hollywood budget.
Yall dreamin'.... go take a look how much sorz cost to generate 30 seconds. Was is something like $100
Wow I've just seen its $500-$700 per 30 secs. 😢
@@armondtanz Won't last long with all the competition. SORA prolly just banking while they can. Have contracts locking in customers as they watch SORA's competitors go dirt cheap. LOL.
Just having a look at the link you added in the description.
Once translated it is worth scrolling down to see the List of Papers that they are publishing.
That's some impressive work right there.
This is the Digital Renaissance. We are here to push innovation. We are looking for semi-realism. I don't care if art schools think my preferred art style is derivative. This is something I want to share with Ai.
I know me & my art can have a verbal conversation. So eat those words in the poetic sense.
Great Video it shows some real progress . I was just telling you about the problems they need the work on like .multiple camera angles and Temporal inconsistency , a few days ago, These are good steps in the right direction
It is really exciting!
I really llike your channel, it's like you test them so we dont have to ,very very good and informative stuff. youve really helped me with assest creation
Thanks, I appreciate that
They want my mobile phone number to register, not in a million years.
Your phone number is public information LOL if you really wanted to try and didn’t trust them with public information for whatever reason then just get google voice and make a few VOIP to signup with so you don’t use your real number directly
That conceptual art piece with televisions is based on the art of Nam June Paik… Would be nice to have that pointed out.
SORA is easily better but this is pretty good too. Open AI is much further ahead then anyone else.
When making stereo shots, make sure the images are highly diffused, like on a overcast day, overcast says make the best 3D shots cause the dimensions of the subject within the shadows remain consistent.. Sunny days are the worst environment for making stereograms. Cause the hard shadows distract the eye from the extra detail, kind of the way a mp3 compressed sound has less dynamics with a loud sound than with sounds that are closer in volume..
Uowww! What a video! Thank you for so much
You're so welcome!
In the first scene of the house by the lake, you will notice that the water is sweeping under the view of the house which is not realisticAs you move closer the water should become more and more revealed, but its moving toward the house.. Or the house is moving toward it. The way to tell if the different camera angles are. consistent is to create stereo imagery of varying IPD (inter pupilary distances), if there is inconsistencies you brain will naturally detect them as warping or inconsistent. Next thing to try is to create a VR180 stereo video, to see if it remains consistent with a skewed fisheye view. That should be easy but it will be easier for the brain to detect inconsistencies.. Another thing to try is to see if you can tile the render, therefore render quadrants of the view . I imagine these can't do this now, but its possible to do it if the frame predictor can direct the render across the tiles to make sure the content remains consistent within each tile, there will also need to be a consistency at the seams. If a tiled render can be achieved with consistency, then it doesn't matter what resolution your renderplane, renders can be distributed across processors for any resolution.
how to sign in to vidu
Epic! Whats up with AiTubo. They seem to be claiming they are using the SORA engine. Please do a review.
Great review
I prefer sora for sure, but its not available for now. Great info presentation. Thank you so much, i needed that badly. Great job, love it👍
Thanks!
Vidu's underwater example doesn't show the effect of compression of the water which would affect the refraction slightly, this would be better observed from a camera outside of the water looking at the dog. However it is modeling the caustics of the light across the dog correctly.. Also one should check to see if the caustics are matched at the surface of the pool.. And if the sunlight shimmers from below, another thing is if the water produces a mirror effects from below against the water , this is due to the difference in density of the air and the water.. Otherwise you could see right through.. However the water will warp the view above the water, and what you can see above the water will be visible toward the center of the view, the perimeter will be reflective. Other considerations is if it can model dirty water with muck floating in the water with correct perspective and at a similar pace. Since these two examples only include some secondary actors, the snow and the water, a more complete test would be to include dirty water and muddy snow.. This is fresh snow which behaves differently from slush.. Fresh snow sticks to each other, a result of the structure of the snowflakes that compose it, they interlock together, and when disrupted disperses in the wind, another thing to check for. Another thing to check for is the collision detection with surfaces that are transparent, like the helmet of a astronaut, will it prevent the head from exiting the helmet or will it pass through, then the training only taught the AI for visible features, not for the transparent features.
Thank you
Didn’t find the button to apply 😂. Interesting what they did.
I noticed something you didn't the reflection on the back of the car is moving up the window, not down the window..
When can we have access to Sora AI and Vidu AI
The shadow below the panda is jumping..
i´ve written a extraordinary fantasy book.... wich nobody understands (complicated cuz of to many characters). cant wait for the day when i am able to produce my own film with ki since this would make my book much more "understandable".
Great series on the AI app comparisons. I think Vidu looks great and that Sora could possibly be left in the dust if they don't watch out. Awesome work. 👍
I'd say sora still seems better
can vidu and sora create a consistent character video?
I created some ai models with haiperai on my shorts. It's mind blowing 😮
GOOD!!
That will be so cool; can we use it yet?
Vidu looks more like Pika Labs or Runway...Not even comparable to Sora.
I disagree, it's a huge step up from Runway
Cant register…. Button is not pressable after filled in name, company, phone etc… anyone succeeded register?
Well, maybe in a few years I'll finally be able to bring my story to life. I really hope so.
Well the fact the homepage is the first one I’ve seen of all the Ai image generators in almost 2 years that’s homepage is all in Chinese = I’ll pass on signing up!
The only ones who win here is us - the users. Yes some dude will find a way to screw sociatiy but that happens every where, science, internetz, and so on.
Almost reminiscent of the difference between DALLE2 and MJ when they were first released.
This is so frustrating. Their sign up list is an insult to my intelligence
Ha! Yes, it is pretty poor UX
Vidu is more realistic than SORA, cause as you said this is of ships in a coffee cup, the waves generated in the coffee cup are much more choppy, water doesn't behave the same at all scales, water at that scale. would be smoother and the boats would suffer from the effect known as surface tension. The Boat in the room is far more realistic because ithe water there is behaving consistent with the scale used.
The link takes me to a site in all Chinese language? Is anyone else experiencing this issue? Is there a way around it for us English speaking folx?
Google Translate
Better tech with Vidu.
Unfortunatly most ai are not free.
You cant fill in the online beta form the send button is greyed out 😭😭😭😭
lmk if that changes. so bummed :/
@@TheIllusionRecords been like that for a while
It's 30 a month.
I thought China wasn't allowed to buy advanced GPU's...
What I think. The world is on the verge of unlocking its advancement.
There was a time when general knowledge was distributed through encyclopedias, exclusively for the very wealthy (remember Barça?).
But that was until the arrival of the internet... and the world has moved faster since then.
Now with A.I. art monopolies are beginning to be broken; drawing, photography, plot, cinema, music... the world speeds up again.
I think it won't be long before the monopolies in the physical industry start to be broken too. The world will be democratically beautiful when its citizens can produce their own goods, such as clothes, cars, appliances, food... and even their medicines.
From that moment on, no one can hold back the advancement of humanity... perhaps with A.I. as a traveling companion.
There is no "monopoly" on those things, what do you mean?
@@kullenberg there's a form of monopoly since me and others can't produce whatever we need/want without a ridiculous amount of money to invest. Right now, if you want clothes or medicines or many other goods you have to buy them from whoever sells them and pay whatever price they want. One day, we could be able to make our own stuff at home, cheap or even free, easy and fully personalized, including medicines. Imagine that future: No more people dying for not being able to pay for meds or dying from side effects of a drug. And that's just one example.
Beautiful future we all dream with. But I bet it will not be even half as good, sadly. Lots of manufacturing materials can be, have been and will get copyrighted so we won't be able to access all the raw materials freely. I see it more realistically as this: you want medicines home printed? You'll have to buy a monthly license to use the patented protein your medicine needs.
Food? Buy a one-use only download of the copyrighted Macdonald's hamburger recipe. Clothes? Download, for a price, the pattern to print the latest jacket from (insert expensive clothing brand here). Extra $10 to get it printed signed by the trendy movie star that made the brand famous.
They'll find ways to cash on it and make the rich get the most of the tech while the poor get the lowest end as usual, if any. I'll assure you that.
And now, if you excuse me, that gives me some ideas for my next movie script, that I hope to animate with AI 👋
@@ronilevarez901 that's beyond ridiculous. You can learn to draw for free with internet access, a pencil and copy paper. No one is gatekeeping that from you, in fact artists are very keen on sharing what they know. There's never been more resources freely available than now.
@@kullenberg All of these areas are under the control of large producers. Even though you are very good at what you do, you will only appear to the general public through them. And guess who gets the lion's share?
Hello! I came across your Udemy course and it's amazing! I'd love to help increase enrollments, generate positive reviews, and drive massive sales. Let's connect and discuss customized strategies. Excited to chat more! Best regards.
China has entered the game. What do you make of VIDU?
Well, I appreciated your video and have already translated their website and requested a call for access. Thank You again.
I think this is just China's propaganda, that i will believe until i get access to the product take EMO for example by bytedance they only examples and didn't given access us now people already forgot about that and that's why i think this just a propaganda of china to there people and for other countries.
Sora is quickly losing first strike advantage.
And 2025 😱😱
Hard to read Chinese
@8:08 walking giant teddy bear with shlong... hmmm, interesting.
Sweet. China can create realistic videos. Nothing bad could come of that, of course, in this era where people are easily influenced. Lol. We are truly at a point in time where you cannot believe something unless you actually see it. Getting kinda wild and awesome, too.
Id believe that there's way more risk of propaganda and fake news dissemination coming from USA's white supremacists who have more than enough money to pay for Sora and similar and whom even USA's government itself is worried about than from a country that barely shares stuff with the rest of the world and whose population has restricted access to the internet.
Listening to you talk makes me tired..