Latent-space scalability for multi-task collaborative intelligence

TL;DR

This paper presents a scalable latent-space approach for multi-task collaborative intelligence, enabling selective decoding for object detection or input reconstruction, optimizing resource use and performance trade-offs.

Contribution

It introduces a novel latent-space scalability method allowing flexible task-specific decoding without reconstructing input pixels, balancing accuracy and efficiency.

Findings

Achieves adjustable performance trade-offs between detection and reconstruction

Reduces computational resources when only detection is needed

Demonstrates superior performance compared to benchmarks

Abstract

We investigate latent-space scalability for multi-task collaborative intelligence, where one of the tasks is object detection and the other is input reconstruction. In our proposed approach, part of the latent space can be selectively decoded to support object detection while the remainder can be decoded when input reconstruction is needed. Such an approach allows reduced computational resources when only object detection is required, and this can be achieved without reconstructing input pixels. By varying the scaling factors of various terms in the training loss function, the system can be trained to achieve various trade-offs between object detection accuracy and input reconstruction quality. Experiments are conducted to demonstrate the adjustable system performance on the two tasks compared to the relevant benchmarks.