It’s a great time to be an AI creator because a new tool seems to come out every week. Sometimes two or three new platforms come out each week. So many tools, so little time.
The new-new in AI video generators is Haiper AI. It came out a few weeks ago and it’s currently in open Beta. I wanted to see what it can do because not all AI-video generators are created the same. So, let’s give it a spin.
Key Features and Functionalities
Haiper currently has a pretty basic setup, which I like. As long as you are somewhat computer savvy, you’ll be able to figure out the UI with minimal interaction.
Text-to-Video: A relatively standard generator. Navigate to the feature from the Explore page by pressing “Create Video With Text,” or go to the Creations page and start typing in the text box at the bottom of the page. The text box gives you several style choices to get you started, or you can create your own. The settings icon allows you to input a seed #, select the aspect ratio, and choose your duration (2 or 4 seconds). Type in your prompt and hit Create to start generating.
Animate Your Image: Have a static image you wish could move? Haiper allows you to upload your images and transform them into lifelike animations. Customize the motion to fit your vision or let Haiper generate its idea, making your content more engaging and visually appealing.
Video Repainting: Sometimes, a fresh look is all a video needs. Haiper’s video repainting tool lets you modify existing videos by altering colors, textures, and elements, giving them a unique and updated appearance. It’s like giving your videos a complete makeover with just a few clicks.
User-Friendly Interface: One of Haiper’s strengths is its intuitive interface. Users can easily navigate through different creation modes, enter text prompts, upload images or videos, and generate content effortlessly. This user-centric design ensures that even those with minimal technical skills can create stunning videos.
Testing…1 2 3
Text-to-Video
I played around with several complex prompts with mixed results. Then I decided to feed it a basic string of prompts to see what I could get, and this produced some fantastic results. Apparently, I was hungry at the time, so my apologies if you haven’t eaten yet. Here are the prompts I gave it:
- A cheeseburger floating in space, ultra-high-definition. (I kept the seed from the initial prompt for all subsequent prompts 1792512680)
- French fries floating in space, ultra-high-definition.
- A milkshake floating in space, ultra-high-definition.
- Chicken nuggets floating in space, ultra-high-definition.
- A chocolate chip cookie floating in space, ultra-high-definition.
- A rack of barbecue ribs floating in space, ultra-high-definition.
- A taco floating in space, ultra-high-definition.
- A slice of pizza floating in space, ultra-high-definition.
- A donut floating in space, ultra-high-definition.
- A deli sandwich floating in space, ultra-high-definition.
I added a quick instrumental track from Suno, and presto, a nice little 20-second video.
The clarity and resolution of the videos are impressive, and the generator did a good job of sticking to the look and feel of the original seed #. I only had one throwaway clip that didn’t look like the original seed video. There was little to no warping on the two-second clips, which is wonderful. On the downside, the generations take 1-2 minutes for a 2-second clip. I’m sure this will improve, but as of now, it’s way too long. I’d also like an option to lock in the seed phrase for each generation, instead of manually copying and pasting each time.
Animate Your Image
Next, I wanted to see how Haiper AI handled animating some images. For this one, I fed it 3 completely different images with varying styles and asked it to generate 4-second clips. I wanted to try more complex images as compared to the relatively simple text-to-video prompts. The images were generated in Leonardo AI. No prompt was given to Haiper to influence its generation. What you see below is the first output Haiper gave me with no regenerating:
Leonardo Image
Haiper 4-Second Clip
Leonardo Image
Haiper 4-Second Clip
Leonardo Image
Haiper 4-Second Clip
The main thing I noticed with animating images is minimal warping. This is a big deal because most image-to-video generators are notorious for severely warping the base image. There was a little warping in example #2, but nothing too crazy. The second thing I noticed is the clarity in the video clips – it’s good. On the downside, I didn’t get too much camera movement such as zooming in or out. But other than that, the clips are high quality and easy on the eyes.
Video Repainting
Let’s take the same three videos that we just animated and see how we can manipulate them with the video repainting tool. Haiper says you can “easily modify colors, textures, and elements to transform your visual content.” Let’s find out.
Haiper Base Video
Action Taken: I selected the skull and told Haiper to make it metallic
Haiper Repainted Video
Haiper Base Video
Action Taken: I selected the creature and told Haiper to make it red, white, and blue.
Haiper Repainted Video
Haiper Base Video
Action Taken: I didn’t select any regions and only selected “Arcane” from the preformatted options.
Haiper Repainted Video
Final Thoughts
In my experience using Haiper, it excelled at creating ultra-high-quality videos using simplistic text-to-video prompts. The videos of the food items floating in space were amazing. Almost zero warping, they look great in full-screen mode, and they all retained a similar look and feel. Haiper also did well at creating videos from images with zero prompting, however, it would have been nice to get a little more camera motion.
I had the hardest time with the repaint feature. I found some success with the skull repaint, but more complex videos weren’t as easy to work with and produced less-than-stellar results. Overall, I think Haiper will be a major player in the AI video creation space. They are actively engaging on Twitter which is a great touch. They replied to my space food video, and it’s cool to know they care about the people using their platform. Well done.
Cheers!
This is a Contributor Post. Opinions expressed here are opinions of the Contributor. Influencive does not endorse or review brands mentioned; does not and cannot investigate relationships with brands, products, and people mentioned and is up to the Contributor to disclose. Contributors, amongst other accounts and articles may be professional fee-based.