Slash Your JavaCV Jar Size by 90%: Maven Platform‑Specific Build Guide

Problem: Why does my project suddenly grow 500 MB?

Many Java developers using JavaCV encounter a huge JAR size increase when they add the javacv-platform dependency.

Result: the built JAR jumps from about 10 MB to over 500 MB, causing massive upload files, disk consumption, and build time swelling from 30 seconds to 10 minutes.

Root cause

javacv-platform

packages native libraries for all platforms (Windows, Linux, macOS, ARM, x86) into the JAR.

Ultimate solution: specify target platform to cut 90% of the size

1. Maven platform‑specific configuration (copy‑paste)

<project>
  <properties>
    <javacpp.platform>windows-x86_64</javacpp.platform>
  </properties>
  <dependencies>
    <dependency>
      <groupId>org.bytedeco</groupId>
      <artifactId>javacv</artifactId>
      <version>1.5.8</version>
    </dependency>
    <dependency>
      <groupId>org.bytedeco</groupId>
      <artifactId>opencv-platform</artifactId>
      <version>4.6.1-1.5.8</version>
    </dependency>
    <dependency>
      <groupId>org.bytedeco</groupId>
      <artifactId>ffmpeg-platform</artifactId>
      <version>5.1.2-1.5.8</version>
    </dependency>
  </dependencies>
  <build>
    <plugins>
      <plugin>
        <groupId>org.bytedeco</groupId>
        <artifactId>javacpp</artifactId>
        <version>1.5.8</version>
        <configuration>
          <properties>
            <property>
              <name>javacpp.platform</name>
              <value>${javacpp.platform}</value>
            </property>
          </properties>
        </configuration>
        <executions>
          <execution>
            <id>process-classes</id>
            <phase>process-classes</phase>
            <goals>
              <goal>build</goal>
            </goals>
          </execution>
        </executions>
      </plugin>
    </plugins>
  </build>
</project>

2. Multi‑platform build with Maven profiles

<profiles>
  <profile>
    <id>windows</id>
    <properties>
      <javacpp.platform>windows-x86_64</javacpp.platform>
    </properties>
  </profile>
  <profile>
    <id>linux</id>
    <properties>
      <javacpp.platform>linux-x86_64</javacpp.platform>
    </properties>
  </profile>
  <profile>
    <id>mac</id>
    <properties>
      <javacpp.platform>macosx-x86_64</javacpp.platform>
    </properties>
  </profile>
</profiles>

Build commands

# Windows platform
mvn clean package -Pwindows

# Linux platform
mvn clean package -Plinux

# macOS platform
mvn clean package -Pmac

Quick command‑line option

mvn clean package -Dmaven.test.skip -Djavacpp.platform=linux-x86_64

Effect comparison

Default platform: 520 MB, 8 minutes build, 15 minutes transfer, massive disk usage.

Specified platform: 48 MB, 1.2 minutes build, 1 minute transfer, comfortable disk usage.

Overall reduction: 90.7 % smaller, build time ↓85 %, transfer time ↓93 %.

Advanced tricks

1. Dynamic platform detection (Java code)

// Java code to load platform‑specific native library
public class PlatformLoader {
    static {
        String os = System.getProperty("os.name").toLowerCase();
        String arch = System.getProperty("os.arch");
        String platform;
        if (os.contains("win")) {
            platform = "windows-x86_64";
        } else if (os.contains("nix") || os.contains("nux")) {
            platform = "linux-x86_64";
        } else if (os.contains("mac")) {
            platform = "macosx-x86_64";
        } else {
            throw new RuntimeException("Unsupported platform");
        }
        System.setProperty("javacpp.platform", platform);
    }
}

2. Docker multi‑stage image

# Multi‑stage Dockerfile to keep only needed native libs
FROM maven:3.8.6-openjdk-11 as builder
WORKDIR /app
COPY pom.xml .
COPY src ./src
RUN mvn clean package -P${TARGET_PLATFORM}

FROM openjdk:11-jre-slim
COPY --from=builder /app/target/app.jar app.jar
ENTRYPOINT ["java", "-jar", "app.jar"]

Common pitfalls

Forgot to set javacpp.platform → all platforms are packaged.

Mixing platform dependency with single libraries → conflicts.

Version mismatch among javacv components → NoClassDefFoundError.

Missing native libraries at runtime → UnsatisfiedLinkError.

“javacv-platform is a developer’s Swiss army knife but a production nightmare; specifying the platform turns a heavyweight into a lightning‑fast app!”

Final recommendation

During development, use the platform artifact for convenience.

In production, specify the target platform to shrink the JAR by ~90 %.

Use Maven profiles in CI/CD for automated multi‑platform builds.