Kling Multi-Cam

Generate new camera angles and shots from your existing video footage. Kling Multi-Cam analyzes your source video and creates complementary shots while maintaining the original cinematic language.

Kling O3 Multi-Cam is the latest generation of Kling's multi-camera technology, replacing the previous O1 version with improved quality and consistency.

What It Does

Kling O3 Multi-Cam creates new camera angles from your existing video:

  • Reverse Angles - Generate the opposite viewpoint of a conversation or scene

  • Alternative Perspectives - Create new shots from different angles

  • Coverage Generation - Fill in missing shots without reshooting

  • Style Continuity - Maintains the look, motion, and feel of your source video

  • Character Consistency - Use Elements to keep characters looking the same across shots

When to Use Multi-Cam

Perfect for:

  • Generating reverse shots for interviews or dialogue scenes

  • Creating coverage when you only shot from one angle

  • Producing alternative perspectives without reshooting

  • Building multi-angle sequences from single-camera footage

  • Visualizing scenes before committing to expensive reshoots

Not ideal for:

How to Use Kling Multi-Cam

  1. Attach your video (3-10 seconds, up to 200MB)

  2. Select "Kling Multi-Cam" from the model selector

  3. Describe the new angle you want to generate

  4. Click send to process

Writing Effective Prompts

Basic Prompt Structure

Tell the AI what shot to generate based on your reference video:

Example Prompts

Reverse Angle (Interview):

"Based on @Video1, generate the reverse angle showing the interviewer. Keep the same lighting and camera style."

Wide Establishing Shot:

"Based on @Video1, generate a wide shot showing the full environment. Maintain the cinematic look and color grade."

Close-Up Coverage:

"Based on @Video1, generate a close-up of the subject's face with the same lighting and mood."

Alternative Perspective:

"Based on @Video1, generate a shot from the left side of the room. Keep the style consistent."

Over-the-Shoulder:

"Based on @Video1, generate an over-the-shoulder shot from behind the subject. Same style and pacing."

Using Elements for Character Consistency

For consistent characters across your generated shots, use Reference Elements:

  1. Create an Element in Library → Elements with 1-4 images of your character

  2. Reference the Element in your prompt with @ElementName

Example with Element:

"Based on @Video1, generate the next shot featuring @Host in a medium shot. Keep the style of the video."

This ensures the character looks the same in your generated coverage as they do in the source video.

Specifications

Setting
Details

Video Duration

3-10 seconds input

Max File Size

200MB

Formats

MP4, MOV

Resolution

720p-2160p input

Output Duration

5 or 10 seconds

Aspect Ratios

16:9, 9:16, 1:1 (auto-detected from source)

Common Workflows

Interview Coverage

  1. Import your single-camera interview clip (10 seconds)

  2. Open in Video Canvas Editor

  3. Select Kling Multi-Cam

  4. Generate reverse angle: "Based on @Video1, generate the reverse angle showing the other person in the conversation"

  5. Generate reaction shot: "Based on @Video1, generate a reaction shot close-up"

  6. Layer with source footage in your edit

B-Roll Variations

  1. Import your hero b-roll shot

  2. Generate wide version: "Based on @Video1, generate a wide establishing shot of this scene"

  3. Generate detail shot: "Based on @Video1, generate a close-up detail shot"

  4. Use all three for dynamic editing

Narrative Scene Coverage

  1. Import your master shot

  2. Create Elements for your characters

  3. Generate medium shots: "Based on @Video1, generate a medium shot of @Character1"

  4. Generate close-ups: "Based on @Video1, generate a close-up of @Character2 reacting"

  5. Build your scene edit with generated coverage

Tips for Best Results

Video Quality

  • Use well-lit, stable source footage

  • Higher resolution inputs produce better outputs

  • Clear, distinct subjects track better

Prompting

  • Be specific about the angle or framing you want

  • Mention "keep the style" or "maintain the look" for consistency

  • Reference specific compositional elements (over-the-shoulder, close-up, wide)

  • Describe the camera position relative to the original

Elements

  • Create Elements for important characters before generating coverage

  • Include 2-4 angles of each character in your Element

  • Name Elements clearly (Host, Guest, ProductShot)

Limitations

  • Duration Range: Source video must be 3-10 seconds

  • AI Generation: Results are AI-generated, not actual footage—may have subtle artifacts

  • Motion Complexity: Very fast or complex motion may not transfer perfectly

  • Character Drift: Without Elements, faces may vary between shots

Troubleshooting

Generated shot doesn't match the style:

  • Add "keep the style of @Video1" or "maintain the cinematic look" to your prompt

  • Try a shorter, more stable section of your source video

Character looks different:

  • Create an Element with 2-4 images of the character from different angles

  • Reference the Element in your prompt with @ElementName

Output quality is low:

  • Ensure source video is at least 720p

  • Use well-lit footage with clear subjects

  • Try a different section of your source video

Generation is taking too long:

  • Multi-Cam can take several minutes to process

  • Check progress in the interface

  • Longer durations (10s) take more time than shorter ones (5s)

  • Kling VFX - Add visual effects to videos

  • Kling Motion Control - Transfer motion to character images

  • VEO Extend - Extend video length by 7 seconds

  • Reference Mode - Character consistency in video generation

Last updated