Make Nature and Landscape Videos With Runway Gen-2

In this guide I'll show you how to make nature and landscape videos both from text and images using the Runway AI platform.

Runway Gen-2 is currently the best AI model for generating videos from text and/or images. Today, I'm going to show you how to make nature and landscape videos both from text and images using the Runway AI platform.

Scenes that feature landscapes can be used in a variety of content, including music videos and commercials. This is the best and most affordable alternative to capturing scenes like this in real life.

If you're new to this platform and AI model, I suggest that you check out my Runway beginner's guide. It's the perfect place to get started with Runway and get the knowledge you need to start making videos.

nature landscape videos using runway ai

How to Make Nature & Landscape Videos With Runway AI

Since you're making videos from text and images, you'll need to use terms popular in cinematography. This is the best way to ensure the videos turn out similar to how you imagined them. This might seem complicated, but it's not. You basically only have to get familiar with certain terms.

You can check out a film glossary to learn some of the terms that are used in cinematography. When applicable, you should mention the angle, camera view, and/or lighting in a video you want to create with Runway. I'll show you exactly what I mean here in some prompt examples later on in the article.

Since you have the option to use only text, images, or a combination of text and images, you're likely going to wonder what the best approach is to creating videos. From what I can tell, using only text prompts can usually create the best results. But image-to-video prompts work great as well.

You can use images that you create in Midjourney, Stable Diffusion or any other text-to-image AI models. You're also free to use any original photographs you have. However, text-only prompts also yield amazing results from time to time.

Finally, you can give vital instructions to the AI model on angles and lighting if you use an image and text combination prompt. If you're new to Runway, this may sound a bit confusing. However, you'll see that it's relatively easy in the next section of the article.

Runway Gen-2 Prompt Examples for Nature & Landscape Videos

I'm going to break down this section into three different categories. The first category revolves solely around text-only prompts. The second features image-to-video prompts. Finally, the third category will show you how to use a combination of text and images to make videos with Runway Gen-2.

Runway Gen-2 Text Prompts for Nature & Landscape Videos

I'll start with basic text prompts that you can use to create beautiful nature and landscape videos. I'll incorporate the tips I shared in the previous section of the article to write quality text prompts. I'm certain that you'll understand the prompt format you should use in Runway to create landscape videos from my examples.

I uploaded the videos to Twitter on an account I have called PromptBase. Let's begin with the prompt examples.

prompt #1: snowy mountain summit, aerial shot showcasing the snow-covered peak, clear crisp daylight, high-angle shot to reveal the landscape below

Wow! I feel like this could be the beginning of an action movie. This is an extraordinary result considering that the video was made from a text prompt. To achieve this, I had to specify the angle, lighting, and camera view. This is a strategy I used for all text prompts in this article.

prompt #2: glacial lake with icebergs, wide-angle shot showcasing the glacier and icebergs, soft diffused light, slow zoom out to reveal the full lake

Let's say that you want to market a perfume with a crisp, clean smell that's associated with snow and cold regions. This is the type of clip you could use in a commercial for a perfume like that. In fact, this clip could be used to promote a variety of products.

prompt #3: coastal cliff sunset, medium shot, capturing the cliff's edge and the sea, golden hour light for a warm, dramatic effect, slightly tilted angle for added depth

Do you need a visualizer for a YouTube video? A short clip that you can loop while music plays in the background can easily be made with Runway AI if you write a good prompt. This is one of those clips that you can use as a visualizer.

prompt #4: wide-angle shot capturing the entire mountain range, soft warm, early morning light, establishing shot from a high vantage point

Many people are sleeping on Runway Gen-2. Although the model has gained a lot of attention, I believe it deserves more. I know this clip doesn't look 100% realistic, but it's absolutely mind-blowing that artificial intelligence managed to create this from two lines of text.

prompt #5: misty forest morning, low-angle shot capturing the forest with mist, soft diffused morning light, slow dolly movement through the misty trees

I could picture this video being used in a commercial. You may have noticed that the first two videos in this article are longer than the others. Don't forget that you can extend the base clip multiple times if you want.

Runway Gen-2 Image Prompts for Nature & Landscape Videos

I'm going to use some images I made in Midjourney for this category of Runway Gen-2 prompts. When you only use the image-to-video model, you don't have to write any text. You only use an image you created with AI or a photograph you took in real life, and the Runway model will do the rest.

I'll write the prompt I used to create each image in Midjourney and then show you how the videos turned out in Runway.

/imagine photorealistic image, aerial shot of a volcano crater lake, sunset with dramatic colors, ultra realistic, full HD, 4k photography --ar 16:9

volcano crater lake landscape midjourney prompt

In my opinion, the prompts you write in Midjourney to generate images that you’ll later use to generate videos in Runway should look similar to the text prompts you write in Runway. You can see the resemblance when you compare the prompt above to those in the previous section of the article. Now that I have a beautiful AI-generated image, I’ll simply upload it to the Runway AI platform. Here’s the result.

There's no doubt that this is a great video, but I honestly prefer the ones that were made using only a text prompt. I feel like you have a lot more control over a video if you use only a text prompt.

/imagine photorealistic image, wide-angle shot of an ice cave, soft icy blue hues coming from the natural ice, the interior of an ice cave, natural light --ar 16:9

ice cave landscape midjourney prompt

I've been to a few ice caves in Iceland. It's something that you have to experience to understand just how glorious it feels to explore a cave that's made entirely out of ice. I've also seen a few clips of parties being held in some of these caves. That's something I have yet to experience, but I know that I'll do it one day. Anyway, let's see how the video turned out from the image above.

I suspect that the tricky part about creating a really good video from an image like this is that there is too much detail. For some areas in the video to look good, you have to sacrifice other areas. The video still looks amazing, but I think future iterations of the Runway image-to-video AI model will be even better for images like this.

/imagine photorealistic image, first person view on a canoe in the middle of a pristine lake, clear midday light, crystal clear water

pristine lake nature landscape midjourney prompt

Midjourney is an amazing AI model for creating photorealistic images, as you can see from the example below. I'm pretty sure I made an image similar to this one at some point in the past. I may have gotten to the point where I have no more original prompt ideas left. Just kidding, but it does feel like that sometimes.

Due to the fact that there's not too much going on in this image, I was able to extend the video two times without noticeably sacrificing the quality. I ended up with a great 12-second clip that could be used for a variety of content.

Runway Gen-2 Combination Prompts for Nature & Landscape Videos

Combination prompts in Runway Gen-2 involve using both an image and text to create videos. You have to be careful how much text you use in these prompts as they can have a huge impact on the output. That’s why I think the best approach is to keep the text part of the prompt very short. You can think about it as giving only one or two extra instructions to the AI model.

Of course, I’ll start by first making an image in Midjourney. I’ll then upload the image to serve as the prompt, and then add text to it.

/imagine photorealistic image, desert dunes under the moonlight, soft cool moonlight, alluring --ar 16:9

desert dunes landscape midjourney prompt

I made this image in Midjourney, and now it's time to turn it into a video. Remember that I said earlier in the article that when you add text to an image prompt, you shouldn't make your instructions too complicated. Instead, try giving a simple direction to the AI model. Essentially, you should use text to describe the scene and how it moves. The only thing I wrote in my text prompt was 'time lapse video'. Here's how it turned out.

The movement of the Moon looks a bit strange in some frames, but that doesn't really matter too much. This is an incredible time-lapse video. There is a very low probability that the Runway Gen-2 AI model would turn the image into a timelapse if I hadn't included the text part of the prompt.

/imagine photorealistic image of a glowing bioluminescent bay, natural bioluminescence, medium shot photograph --ar 16:9

bioluminescent bay midjourney prompt

In case you didn't know, there are several popular bioluminescent beaches around the world. They look very similar to the one in the image above. Bioluminescent microorganisms are what make the water glow. For the text part in this combination prompt, I wrote 'slow tracking shots in the bay's waters'.

The waves are not powerful enough to make rocks move, yet this still occurs in the video above. That's completely okay because this AI model is still in its early stages. It will get much better over time. Until then, we can still appreciate the beautiful videos it can produce. Apart from the big rock moving on the left side of the scene, I have to say that this video looks amazing considering that it’s generated with artificial intelligence.

Final Thoughts

The Runway AI platform provides users with numerous options for creating videos. You have complete freedom to choose any of the three approaches I talked about in the previous section of the article. Personally, what I like the most is to write a really good text prompt and create a video from it.

The main thing to pay attention to when you’re writing text-only prompts in Runway is that you should describe the scene similarly to how a cinematographer would. If you’re not familiar with the terms used in cinematography, you can check out a film glossary online. You don’t need to learn every single term, but only the ones that relate to the angle, camera view, and lighting.

Before I finish this article, let's just take a second to appreciate how far technology has come. Just imagine what it would take to make some of these videos in real life. For instance, let's say that you wanted an aerial shot of a mountain range.

You'd have to physically travel to the mountains, have at least a good drone set up, and great weather conditions to record videos. With Runway AI, you can generate a video in less than a minute on your smartphone or computer. I can't wait to see what's next in the AI video generation space.

Leave a Comment