NSFW Video Detection API: AI-Powered Content Moderation

The NSFW (Not Safe For Work) Video Content Detection API is an AI-driven tool designed to automatically scan video files and identify inappropriate content. This API is particularly useful for platforms that handle user-generated video content, ensuring compliance with community standards and maintaining a safe user environment. By leveraging advanced artificial intelligence models, the API provides rapid and accurate detection of explicit or unsafe material, making it an essential component for content moderation strategies.

The NSFW Video Content Detection API boasts several key features that make it a powerful tool for content moderation: * **Rapid Video Scanning:** Utilizes advanced AI models to quickly scan video content for NSFW material. * **Adjustable Safety Tolerance:** Allows users to fine-tune the strictness levels, providing flexibility in defining what is considered inappropriate. * **Scalable Asynchronous Processing:** Supports large-scale video uploads with asynchronous processing, ensuring efficient handling of frequent video submissions. * **Clear Status Indicators:** Delivers results with clear status indicators, facilitating straightforward integration into existing moderation workflows. * **Comprehensive Status Lifecycle:** Offers a robust handling of asynchronous tasks with a detailed status lifecycle (starting → processing → succeeded/failed).

The applications of the NSFW Video Content Detection API are diverse and impactful across various industries: * **Moderation for Social Media:** Automatically screen user-uploaded videos for explicit or sexual content before they are published, ensuring a safe and compliant platform. * **Video Hosting Platforms:** Perform large-scale scanning of incoming video streams to maintain compliance with community standards and prevent the spread of inappropriate content. * **Adult Content Filters:** Filter out NSFW material in corporate or educational settings to maintain a safe and professional user environment. * **Advertising Platforms:** Prevent brand-unsafe content from appearing in video advertisements, protecting brand reputation and ensuring compliance with advertising guidelines. * **Automated Monitoring:** Continuously audit archived video libraries to identify and remove non-compliant clips, ensuring ongoing adherence to content standards.

To effectively use the NSFW Video Content Detection API, it's crucial to understand its endpoints and authentication methods: * **Base URL:** All requests should be sent to `https://prod.api.market/api/v1/magicapi/nsfw-video-content-detection`. * **Authentication:** Include your API key with every request using the header `x-magicapi-key: YOUR_API_MARKET_KEY`. * **Asynchronous Processing:** The API processes videos asynchronously, requiring a two-step process: 1. Send a POST request with the video URL to create a prediction. 2. Poll the GET endpoint with the prediction ID to check the status until it is either “succeeded” or “failed”. * **Endpoints:** * **Create a Prediction (POST):** `/api/v1/magicapi/nsfw-video-content-detection/predictions` * **Get Prediction Status and Results (GET):** `/api/v1/magicapi/nsfw-video-content-detection/predictions/{id}`

Implementing the NSFW Video Content Detection API involves sending requests to the API endpoints and handling the responses. Here are examples in JavaScript, Python, and cURL: **JavaScript (Node.js):** ```javascript const apiKey = "YOUR_API_MARKET_KEY"; const baseUrl = "https://prod.api.market/api/v1/magicapi/nsfw-video-content-detection"; async function createAndWaitForPrediction(inputData) { try { // Step 1: Create the prediction const response = await fetch(`${baseUrl}/predictions`, { method: "POST", headers: { "Content-Type": "application/json", "x-magicapi-key": apiKey }, body: JSON.stringify({ version: "ef94e6bbf329eebc092601a647996431056a58f479d0f4129ea6e8171374d158", input: inputData }) }); if (!response.ok) { throw new Error(`API error: ${response.status} ${response.statusText}`); } const prediction = await response.json(); const predictionId = prediction.id; // Step 2: Poll for results with timeout return await pollForResults(predictionId); } catch (error) { console.error("Error:", error); throw error; } } function delay(ms) { return new Promise(resolve => setTimeout(resolve, ms)); } async function pollForResults(predictionId) { const MAX_TIMEOUT = 600000; // 10 minutes const POLL_INTERVAL = 1000; // 1 second const startTime = Date.now(); while (Date.now() - startTime < MAX_TIMEOUT) { const response = await fetch(`${baseUrl}/predictions/${predictionId}`, { headers: { "x-magicapi-key": apiKey } }); if (!response.ok) { throw new Error(`API error: ${response.status} ${response.statusText}`); } const prediction = await response.json(); if (prediction.status === "succeeded") { return prediction.output; } else if (prediction.status === "failed") { throw new Error(`Prediction failed: ${prediction.error || "Unknown error"}`); } await delay(POLL_INTERVAL); } throw new Error("Prediction timed out after 10 minutes"); } // Usage Example: (async () => { const input = { video: "https://replicate.delivery/pbxt/MeYrKketDRwUdUPYlHEsA0UlcD4eOlFegxJwvJzFuhL1en1O/falcon2.mp4", safety_tolerance: 2 }; try { const result = await createAndWaitForPrediction(input); console.log("Prediction output:", result); } catch (error) { console.error("Prediction error:", error); } })(); ``` **Python:** ```python import requests import time API_KEY = "YOUR_API_MARKET_KEY" BASE_URL = "https://prod.api.market/api/v1/magicapi/nsfw-video-content-detection" def create_and_wait_for_prediction(input_data): # Step 1: Create the prediction headers = { "Content-Type": "application/json", "x-magicapi-key": API_KEY } response = requests.post( f"{BASE_URL}/predictions", headers=headers, json={ "version": "ef94e6bbf329eebc092601a647996431056a58f479d0f4129ea6e8171374d158", "input": input_data } ) if response.status_code not in [200, 201]: raise Exception(f"API error: {response.status_code} {response.text}") prediction = response.json() prediction_id = prediction["id"] # Step 2: Poll for results with timeout return poll_for_results(prediction_id) def poll_for_results(prediction_id): MAX_TIMEOUT = 600 # 10 minutes POLL_INTERVAL = 1 # 1 second headers = {"x-magicapi-key": API_KEY} start_time = time.time() while time.time() - start_time < MAX_TIMEOUT: response = requests.get( f"{BASE_URL}/predictions/{prediction_id}", headers=headers ) if response.status_code != 200: raise Exception(f"API error: {response.status_code} {response.text}") prediction = response.json() if prediction["status"] == "succeeded": return prediction["output"] elif prediction["status"] == "failed": error_msg = prediction.get("error", "Unknown error") raise Exception(f"Prediction failed: {error_msg}") time.sleep(POLL_INTERVAL) raise Exception("Prediction timed out after 10 minutes") # Usage: if __name__ == "__main__": input_data = { "video": "https://replicate.delivery/pbxt/MeYrKketDRwUdUPYlHEsA0UlcD4eOlFegxJwvJzFuhL1en1O/falcon2.mp4", "safety_tolerance": 2 } try: result = create_and_wait_for_prediction(input_data) print("Prediction output:", result) except Exception as e: print("Error:", e) ``` **cURL:** ```curl # Step 1: Create a prediction curl -X POST "https://prod.api.market/api/v1/magicapi/nsfw-video-content-detection/predictions" \ -H "Content-Type: application/json" \ -H "x-magicapi-key: YOUR_API_MARKET_KEY" \ -d '{ "version": "ef94e6bbf329eebc092601a647996431056a58f479d0f4129ea6e8171374d158", "input": { "video": "https://replicate.delivery/pbxt/MeYrKketDRwUdUPYlHEsA0UlcD4eOlFegxJwvJzFuhL1en1O/falcon2.mp4", "safety_tolerance": 2 } }' # Sample response: # { # "id": "ufawqhfynnddngldkgtslldrkq", # "status": "starting", # ... # } # Step 2: Check prediction status (replace with the actual prediction ID) curl -X GET "https://prod.api.market/api/v1/magicapi/nsfw-video-content-detection/predictions/ufawqhfynnddngldkgtslldrkq" \ -H "x-magicapi-key: YOUR_API_MARKET_KEY" ```

Understanding the pricing and usage guidelines is essential for cost-effective implementation: * **API Unit:** 1 API unit corresponds to processing 1 video. * **Free GET Requests:** GET requests to retrieve prediction status are free and do not count toward API unit usage. * **Pricing Tiers:** Various pricing tiers are available based on your monthly video processing volume. It’s important to choose a pricing tier that aligns with your usage needs to optimize costs.

To maximize the efficiency of the NSFW Video Content Detection API, consider the following performance optimization tips: * **Batch Requests:** Where possible, batch your requests to reduce overhead. * **Efficient Polling:** Use efficient polling intervals (1-second intervals generally suffice) to avoid unnecessary requests. * **Upgrade Plan:** If you frequently need faster results or are processing large volumes, consider upgrading your plan or distributing requests across multiple workflows.

Here are some frequently asked questions about the NSFW Video Content Detection API: * **How do I handle very large video files?** * Upload them to a secure location and provide the direct link as “video.” Ensure your server can stream them reliably. * **Which video formats are supported?** * Most common video formats (e.g., MP4) are supported as long as they can be hosted and accessed via HTTPS. * **Can I adjust the sensitivity?** * Yes, use the “safety_tolerance” parameter (1 is strictest, 6 is most permissive). * **Do GET requests cost API units to check the status?** * No, GET status checks are free. * **What if the video is partially NSFW?** * The model processes the entire video. If any segment is flagged as NSFW, the final output indicates NSFW content.