Rethinking Implicit Neural Representations for Vision Learners
This work addresses a gap in applying INRs to high-level vision tasks, which could benefit researchers and practitioners in computer vision, though it appears incremental as it builds on existing INR concepts.
The authors tackled the problem of extending Implicit Neural Representations (INRs) from low-level to high-level vision tasks by reformulating INR definitions and proposing an Implicit Neural Representation Network (INRN), achieving effectiveness in tasks like image classification, object detection, and instance segmentation.
Implicit Neural Representations (INRs) are powerful to parameterize continuous signals in computer vision. However, almost all INRs methods are limited to low-level tasks, e.g., image/video compression, super-resolution, and image generation. The questions on how to explore INRs to high-level tasks and deep networks are still under-explored. Existing INRs methods suffer from two problems: 1) narrow theoretical definitions of INRs are inapplicable to high-level tasks; 2) lack of representation capabilities to deep networks. Motivated by the above facts, we reformulate the definitions of INRs from a novel perspective and propose an innovative Implicit Neural Representation Network (INRN), which is the first study of INRs to tackle both low-level and high-level tasks. Specifically, we present three key designs for basic blocks in INRN along with two different stacking ways and corresponding loss functions. Extensive experiments with analysis on both low-level tasks (image fitting) and high-level vision tasks (image classification, object detection, instance segmentation) demonstrate the effectiveness of the proposed method.