Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > OpenKatana > OpenKatana · Analyze reference

ComfyUI Node: OpenKatana · Analyze reference

Class Name

OpenKatana_analyze_reference

Category
OpenKatana/Tools
Author
rk0neki (Account age: 235days)
Extension
OpenKatana
Latest Updated
2026-11-05
Github Stars
0.05K

How to Install OpenKatana

Install this extension via the ComfyUI Manager by searching for OpenKatana
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter OpenKatana in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

OpenKatana · Analyze reference Description

Analyzes reference video editing, cuts, pacing, color, motion, and transitions at 60 fps.

OpenKatana · Analyze reference:

OpenKatana_analyze_reference helps you study how a reference video is edited so you can make more informed creative decisions about timing, pacing, visual style, and transitions. Its frame-level measurement system examines references on a fixed 60-frames-per-second clock, allowing it to capture short events such as flashes, one-frame cuts, rapid movements, and changes in transition timing. Available measurements cover shot cuts, repeated-frame cadence, tonal and color characteristics, the persistence of small bright textures, and motion, blur, and brightness changes immediately after cuts. Cut detection uses brightness-normalized image comparisons and distinguishes real cuts from temporary changes where the original picture returns shortly afterward, which are reported as dips. These measurements help you understand a reference rather than automatically reconstruct it: they are estimates you should verify visually, especially when interpreting complex transitions or subtle texture effects.

OpenKatana · Analyze reference Input Parameters:

project_id

A required STRING that identifies the OpenKatana project associated with the analysis request. This lets the tool operate in the context of the intended project rather than treating the request as an isolated operation. The supplied API test creates a project before running reference analysis, so you should use an identifier for a project that already exists in your OpenKatana workspace. No default value, allowed identifier pattern, minimum length, or maximum length is specified in the supplied node interface.

arguments

A required STRING containing the tool arguments used to configure the reference-analysis request. This is where you supply the analysis details rather than adding extra input sockets to this node. The supporting context shows reference analysis using a source video, a start time, a duration, a threshold, and a selection of measurements; the documented measurement names are cuts, cadence, look, textures, and transitions. These choices determine which part of the reference is examined and which aspects of its editing are measured. Use the argument structure accepted by your installed OpenKatana tool, typically a JSON object encoded as text; the supplied wrapper does not expose the argument parser or complete tool schema, so exact field requirements, threshold behavior, defaults, and numerical limits cannot be confirmed from this interface. No default value or string-length limits are shown for this input.

after

An optional STRING accepted by the node's execution method. Its name suggests an execution-order or dependency-related role, but the supplied implementation does not read it inside the execution body or explicitly pass it to the tool invocation. Consequently, the provided code does not establish that this value changes the reference measurements or guarantees that another operation finishes first. Leave it unconnected unless your surrounding OpenKatana workflow documents how to use it. Its Python-level default is None when omitted; no STRING default, allowed values, or length limits are shown in the supplied interface.

OpenKatana · Analyze reference Output Parameters:

result_json

A STRING containing the tool response in JSON form, intended for inspection or processing by later workflow steps. The supporting API test confirms that a reference-analysis response can contain a measured object with a cuts list, where entries include a frame value and a kind value; entries whose kind is cut represent detected shot changes, while the measurement documentation also describes temporary changes reported as dip. Reference measurement frame numbers use the fixed 60-fps analysis clock, not necessarily the source video's native frame rate. Depending on the requested measurements, the response can also describe repeated pictures and unique-picture rate, tonal and color characteristics, bright-speck persistence, or post-cut transition behavior; transition measurements are documented as including local_curve arrays. Treat these values as evidence to review rather than a ready-made recreation of the reference. The complete response schema and any additional wrapper fields are not shown in the supplied code.

OpenKatana · Analyze reference Usage Tips:

  • Start with a short, representative section of your reference by configuring its start time and duration in the tool arguments. This makes it easier to compare the measurements with visible events before examining a longer edit; the supplied test analyzes a three-second section.
  • Request cuts when your main goal is understanding edit pacing. When reading the result, filter for entries whose kind is cut rather than treating every reported event as a new shot, because temporary picture changes can be classified as dips.
  • Interpret frame numbers on a 60-fps clock. For example, analysis frame 120 corresponds to two seconds on that clock; for a section starting partway through a video, confirm the response's frame-origin convention before translating it into absolute source time.
  • Use cadence to understand repeated frames before judging motion smoothness. Lower-frame-rate material is represented by repeated pictures on the 60-fps clock; 24-fps material can show a 2–3 repetition pattern, which does not mean the analysis has created new motion detail.
  • Use look for studying tone, saturation, and shadow or highlight color casts, and textures for studying how long small bright specks persist. Texture measurement concerns the overlay's apparent picture rate, so do not assume it is the same as the underlying footage's cadence.
  • Use transitions to study common behavior immediately after real cuts. Its curves are aggregated as a median across cuts to reduce the influence of individual camera moves, so check unusual transitions separately rather than assuming the summary represents every cut exactly.
  • Do not assume a universal threshold setting. The supplied test uses a threshold of 0.25, but the provided code does not establish its allowed range or how increasing it affects sensitivity; check your installed tool documentation and validate the result against known cuts.
  • Ensure the OpenKatana environment has the vision extra and can run ffprobe before starting analysis. Use a source that the same runtime can access and that contains an actual video stream.

OpenKatana · Analyze reference Common Errors and Solutions:

Install OpenKatana with the vision extra to measure a reference

  • Explanation: This is an explicit error in the supplied measurement code. The analysis requires OpenCV, and its cv2 module could not be imported.
  • Solution: Install OpenKatana with its vision extra in the Python environment running ComfyUI, following the installation instructions for your OpenKatana version, then restart ComfyUI.

The reference has no video stream to measure

  • Explanation: This explicit error occurs when the video-probing step finds no video stream. The selected source may be audio-only, unreadable, damaged, or otherwise unavailable to the analysis environment.
  • Solution: Confirm that the source plays as video, check its path from the ComfyUI runtime, and try a known-good video file. If needed, re-export the reference into a format supported by your local FFmpeg installation.

unexpected inputs: ...

  • Explanation: The node explicitly rejects extra named fields that are not part of its declared tool fields. The supplied interface exposes only project_id, arguments, and optional after; analysis settings are not additional sockets on this node.
  • Solution: Remove unsupported fields from custom calls or workflow modifications. Put the reference-analysis settings inside arguments using the installed tool's accepted argument schema.

Invalid or malformed arguments — exact message depends on the argument parser

  • Explanation: The node processes arguments before invoking the tool. Incorrectly formatted text or unsupported argument names can prevent that request from being constructed, although the supplied code does not show the parser's exact error messages.
  • Solution: Check the argument format against your installed OpenKatana schema. If JSON is required, use double-quoted keys and strings, avoid trailing commas, and pass numbers as numbers rather than quoted text. Begin with the smallest valid request before adding measurement options.

Project not found or unavailable — exact message is not shown

  • Explanation: Reference analysis is called with a project identifier, and the supplied test creates the project before invoking it. A missing project or an identifier from a different workspace can therefore be a project-context problem, though the exact validation behavior is not provided.
  • Solution: Create or select the intended project first, copy its identifier exactly, and confirm that your ComfyUI instance is using the workspace containing that project.

ffprobe could not be started — operating-system message may vary

  • Explanation: The measurement code launches ffprobe as an external program. If it is not installed or is not discoverable in the runtime's executable search path, video inspection cannot begin.
  • Solution: Install an FFmpeg distribution that includes ffprobe, make it available to the environment running ComfyUI, and restart ComfyUI after changing the executable search path.

Video probing timed out — exception wording may vary

  • Explanation: The supplied probing operation has a 60-second timeout. A slow, inaccessible, or problematic source can prevent video information from being read within that time.
  • Solution: Copy the reference to a reliable local location, verify that it opens normally, and retry. If the problem persists, re-export the source and test the new file.

Measured frames or transitions do not match expectations

  • Explanation: This is an interpretation or estimation issue rather than a documented exception. Native source frames can differ from the 60-fps analysis clock, dips are distinct from cuts, and transition curves summarize behavior across multiple cuts rather than describing each transition independently.
  • Solution: Review the corresponding reference frames visually, interpret timing at 60 fps, separate cut entries from dip entries, and analyze a shorter section when a particular transition needs closer inspection.

OpenKatana · Analyze reference Related Nodes

Go back to the extension to check out more related nodes.
OpenKatana
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

OpenKatana · Analyze reference