CVMar 5, 2024

Deep Common Feature Mining for Efficient Video Semantic Segmentation

arXiv:2403.02689v26 citationsh-index: 4Has CodeIEEE transactions on circuits and systems for video technology (Print)
Originality Incremental advance
AI Analysis

This addresses efficiency and reliability issues in video semantic segmentation for computer vision applications, representing an incremental improvement.

The paper tackles redundant computation and unreliable feature propagation in video semantic segmentation by introducing Deep Common Feature Mining (DCFM), which decomposes features into common and independent components for efficient reuse, achieving a superior balance between accuracy and efficiency on VSPW and Cityscapes datasets.

Recent advancements in video semantic segmentation have made substantial progress by exploiting temporal correlations. Nevertheless, persistent challenges, including redundant computation and the reliability of the feature propagation process, underscore the need for further innovation. In response, we present Deep Common Feature Mining (DCFM), a novel approach strategically designed to address these challenges by leveraging the concept of feature sharing. DCFM explicitly decomposes features into two complementary components. The common representation extracted from a key-frame furnishes essential high-level information to neighboring non-key frames, allowing for direct re-utilization without feature propagation. Simultaneously, the independent feature, derived from each video frame, captures rapidly changing information, providing frame-specific clues crucial for segmentation. To achieve such decomposition, we employ a symmetric training strategy tailored for sparsely annotated data, empowering the backbone to learn a robust high-level representation enriched with common information. Additionally, we incorporate a self-supervised loss function to reinforce intra-class feature similarity and enhance temporal consistency. Experimental evaluations on the VSPW and Cityscapes datasets demonstrate the effectiveness of our method, showing a superior balance between accuracy and efficiency. The implementation is available at https://github.com/BUAAHugeGun/DCFM.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes