Haiper AI, Try This FREE Next-level AI VIDEO

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
hyper AI is a brand new AI video generator that is completely free and available so if you enjoyed the videos that came out of Sora well you can make your own today I'm going to take you through some of the examples coming out of hyper show you how to use it and much much more so let's explore the latest Frontier in AI video together just take a look at these videos we get some fantastically crisp realistic video videos showing a huge variety of different scenes with excellent animation coherence and quality to all of their videos now if we look at a realistic scene of a car splashing water and you can really sense the spray I love these effects of drops almost hitting the lens now in this example we see a lovely delicate collection of carrots slowly floating down a river and just take a look at the reflections on the water but we can take a look at some more photorealistic interpretations especially of this woman you can see that she looks uh relatively real I would say the only issue is that her lips are a little bit fake like little slugs on her face but this is a accurate representation of Elon Musk and the thing I particularly love about this AI tool is the stylistic intent that is obvious the Artistry the aesthetic that naturally comes out of these videos that they all look particularly visually engaging which is incredible that the colors the composition the tones and the taste of the works are incredible look at this image of an eye each individual eyelash has been carefully rendered and I believe that if you take a look at this you would be pretty shocked that it was not a real video now the best thing is is is that it's completely accessible and completely free right now so you can come to hyper click try for free and immediately you'll be led into the platform where you can enter in any Simple Text prompt that you like and generate your own realistic AI video in seconds now to do this all you have to do is come down to the text box at the bottom of the page and describe what you would like to see for this I simply recommend that you add in a subject a type of motion and a style so they're three things that we're looking at so I asked for a beautiful blonde Swedish woman walking towards the camera blurred background City Scene Tokyo night and you can go ahead and click on Create and what's great about hyper is the ease of use so you can simply enter a very basic prompt like this and immediately get out a hyper realistic video now I'll tell you a little bit more about hyper whilst this is loading it's been developed by two ex employees of Deep Mind and deep mind is a Google owned artificial intelligence company with a great background in AI Tech they've raised around 13 million in seed funding they originally aimed to create a tool for 3D reconstruction which is creating 3D models using AI but they switched to video generation due to it being a more fascinating problem now there is hope that this may well remain free for a little while as Mio said that the company aims to keep these features free in order to build a community he noted that it is too early in the startups journey to think about building a subscription product around video generation here is the Swedish woman walking through Tokyo and you can see it's absolutely incredible that the individual strains of hair are being animated and the cars are moving coherently in the background we've got a wonderful cinematic bcka effect here and everything about the individual is anatomically correct however I would say that perhaps it has not accurately created the movement that I've requested which is for the protagonist to be walking towards the camera it's almost as if she's throwing a look back over her shoulders seductively now if you want to download your work all you have to do is press on the download icon underneath each image and this vertical video was exported at 720 by 1088 whereas the horizontal videos are coming out at 1280 by 720 so we're getting a high definition of video coming out and it's also Poss to then take these and upscale them further using a tool like topaz's video now hyper gives a couple of pointers for Effective prompting inside of its tool in general it's good to use clear and specific language provide detailed descriptions of the subject and scene to help the AI model generate accurate images make sure to be concise use concise language avoiding unnecessary vocabulary that might confuse the model or dilute the intended meaning meaning finally be creative employ interesting vocabulary metaphors or contrasts to stimulate the model's imagination and creativity you can see here they're really anthropomorphizing their tool they're believing that it has in its own imagination and creativity and essentially we're getting pretty close now one thing that I particularly appreciate is that when you are describing a scene you have the ability to add in a stylistic tag so they have a number of different styles which will automatic apply a type of feeling and visual influence onto your work and the ones they have available here are old film watercolor cyber Punk Arcane Arcane means mysterious Lego blurred background which is a bocker effect and generally gives you a very cinematic feel as you can see in my video here I've actually blurred out this plant over my shoulder and if I I do this hopefully it will blur out more of the content so you can use these as a St starting point to really stylize your work now there is also an i available here which allows you to either create in public or turn it off and allow your Creations to be kept private next up you notice there are these two little candy icons on their side two little sliders and this essentially gives us more parameters to play around with at the moment you can only create 2 second long videos which is quite short but the quality is so high I am completely willing to overlook that fact and they have mentioned that they will be expanding the length of these videos before too long after that there's also the seed and what the seed does is it is the starting point for generating the initial image grid a seed is randomly generated for each image but can be specified using parameters if you use the same seed number and prompt you will get similar final videos so if you want to recreate a video and keep it as similar as possible each time you generate it and just slightly tweak a few things well then you can use seed now there are a number of different ways to create your videos inside of hyper and to explore these you can come to the green circle icon on the left hand side of the screen and from here you can select from create with text which is essentially using a prompt you can use animate your image which is where you upload an image and it will use it as a base to then animate from this is f fantastic for taking your AI artwork from other platforms like mid Journey or ideogram and giving them a animated flavor so here is a image that I uploaded to hyper and asked it to animate for me now there is also a videoo video tool which promises to allow you to change elements inside of your video this is incredibly exciting and it gives us the power to take existing videos and change them now let's take a look at how this matches up to Sora we can see using this same prompt that we get some woolly mammoths walking towards the camera I would say that the Sora model is still more realistic but comparing this to the existing models that we have it's absolutely incredible let's take another look at a SORA comparison this is the same prompt this one is coming out of our dear friend hyper and here we can look at the prompt that was put into Sora and you can see that the Sora version I would say is still winning out just but it's incredibly close now hyper is creating a number of creative applications for this tool but I particularly like their idea of taking sketches and turning them into videos here you can see they've turned this simple image into this stunning rendering of this rather aggressive looking bunny rabbit I would not like to to meet him on a battlefield 2024 is certainly shaping up to being the year of AI video with tools like Runway which is backed by Nvidia and Google and also open AI who have teased and tempted with their Sora model are really ramping up the game and we're seeing huge rapid advancements to the technology as Rebecca hunt a partner at octopus Ventures said once the models get to point that transcends The Uncanny Valley and reflects the real world and all its physics there will be a period where the applications are infinite the implications and applications of AI video are immensely exponentially more impactful than that of AI art of still images we're going to see whole worlds whole narratives whole films being created with simple text prompts we've seen recently with products like LTX Studio and morph Studio the potential for taking AI video and being able to simultaneously adjust a number of Clips in one go so if you are creating an entire advert and you want to change the color of your product across all shots you can do that in an instant and it's this further degree of complexity and production that we're going to see start to evolve out of the latest AI tools and Technologies and what I'm excited about is as we're going to start to develop AI video we're also going to see the ability for multimodal AI input so we'll be able to ask the AI to generate not just the video but actually come up with story lines with dialogue with music and voice and synthesize all of these together whilst maintaining consistent characters consistent voice tones vocal inflections and stylistic intent it's an incredibly exciting time I mean slightly giddy with excitement I'm feeling a little bit lightheaded here oh there's so many AI announcements how does one stay a breast well if you want to stay a breast I do do my best to give you sincere truthful and honest content on the latest AI tools and if you would like to stay with me do consider touching this button down here I believe there's a subscription button somewhere around here let me know if you enjoyed this video And subscribe if you want to stay updated with some of the latest AI news I'm Samson Vols and I hope you have a delightful day
Info
Channel: AI Samson
Views: 92,846
Rating: undefined out of 5
Keywords: ai, samson vowles, haiper, sora ai, haiper tutorial
Id: RETrzlePRys
Channel Id: undefined
Length: 11min 43sec (703 seconds)
Published: Thu Mar 07 2024
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.