Research map / Detection and segmentation / CNN features
Towards Effective Waste Segmentation for Automated Waste Recycling in Cluttered Background
Mamoona Javaid, Mubashir Noman, Abdul Hannan, Shah Nawaz, Mustansar Fiaz, Sajid Ghuffar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SCM to baseline (ablation) | EWSegNet baseline · ZeroWaste-f test | mIoU (%): 47.32 -> 51.63 | Improved |
| add FCM to baseline (ablation) | EWSegNet baseline · ZeroWaste-f test | mIoU (%): 47.32 -> 53.05 | Improved |
| add SCM to FCM (ablation) | EWSegNet · ZeroWaste-f test | mIoU (%): 53.05 -> 54.11 | Improved |
| add AFEM to FCM + SCM (ablation) | EWSegNet · ZeroWaste-f test | mIoU (%): 54.11 -> 56.44 | Improved |
| add AFEM to FCM + SCM (ablation) | EWSegNet · ZeroWaste-f test | pixel accuracy (%): 91.72 -> 91.75 | No clear effect |
| raise initial learning rate from 5e-5 to 1e-4 (ablation) | EWSegNet · ZeroWaste-f test | mIoU (%): 56.44 -> 57.14 | Improved |
| raise initial learning rate from 5e-5 to 1e-4 (ablation) | EWSegNet · ZeroWaste-f test | Metal IoU (%): 35.05 -> 38.78 | Improved |
| use EWSegNet instead of COSNet | encoder · ZeroWaste-f test | mIoU (%): 56.67 -> 56.44 | No clear effect |
| use EWSegNet instead of COSNet | encoder · ZeroWaste-f test | latency (msec): 73.6 -> 64.8 | Improved |
| use EWSegNet instead of LWCHNet | segmentation network · ZeroWaste-aug test | mIoU (%): 63.16 -> 74.10 | Improved |
| use EWSegNet instead of COSNet | segmentation network · SpectralWaste test | mIoU (%): 69.96 -> 71.03 | Improved |
| use EWSegNet instead of COSNet on rigid plastic | segmentation network · ZeroWaste-f test | Rigid Plastic IoU (%): 37.24 -> 33.28 | Hurt |
Closest papers by meaning
- UrbanWaste: In-the-Bin Dataset for Waste Disposal Inspection with Multi-Granularity Hierarchical Labels (AAAI 2025 · similarity 0.89)
- ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenes (CVPR 2022 · similarity 0.89)
- WISNet: Pseudo Label Generation on Unbalanced and Patch Annotated Waste Images (CVPR 2025 · similarity 0.87)
- ETHSeg: An Amodel Instance Segmentation Network and a Real-World Dataset for X-Ray Waste Inspection (CVPR 2022 · similarity 0.84)
- Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling (CVPR 2024 · similarity 0.78)
- Reinforcement Learning as a Parsimonious Alternative to Prediction Cascades: A Case Study on Image Segmentation (AAAI 2024 · similarity 0.77)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
