Research map / Architectures and efficiency / Quantization
Bandwidth-Aware Adaptive Codec for DNN Inference Offloading in IoT
Xiufeng Xie, Ning Zhou, Wentao Zhu, Ji Liu
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 YUV tuning (ablation) | AutoJPEG · Cityscapes (DRN-D-38) | mIoU (%): 70.94 -> 71.24 | Improved |
| add YUV tuning (with DNN tuning) (ablation) | AutoJPEG · Cityscapes (DRN-D-38) | mIoU (%): 71.17 -> 71.24 | Improved |
| add DNN tuning (ablation) | AutoJPEG · Cityscapes (DRN-D-38) | mIoU (%): 70.75 -> 70.94 | Improved |
| add both DNN and YUV tuning (ablation) | AutoJPEG · Cityscapes (DRN-D-38) | mIoU (%): 70.75 -> 71.24 | Improved |
| replace GRACE with AutoJPEG | DRN-D-38 · Cityscapes | mIoU (%): 0.24% lower mIoU at 418KB vs 408KB | Improved |
| replace JPEG 85% with AutoJPEG at similar size | DRN-D-22 · Cityscapes | mIoU loss (%): 7.43 -> 0.26 | Improved |
| replace JPEG 100% with AutoJPEG | DRN-D-22 · Cityscapes | mIoU loss (%): 0.69 -> 0.15 | Improved |
| finetune DNN on JPEG-encoded images | JPEG · Cityscapes | mIoU | Improved |
| replace JPEG75%+finetune with AutoJPEG | ResNet56 · CIFAR10 | Top-1 accuracy (%): +4.73 | Improved |
| replace Neurosurgeon-Prune with AutoJPEG | DRN-D-38 · Cityscapes | mIoU: more than 2% | Improved |
Closest papers by meaning
- Ultra-Lightweight Adaptive Bitrate Deep Video Compression (IJCV 2026 · similarity 0.87)
- Partial Weight Adaptation for Robust DNN Inference (CVPR 2020 · similarity 0.86)
- ECC: Platform-Independent Energy-Constrained Deep Neural Network Compression via a Bilinear Regression Model (CVPR 2019 · similarity 0.85)
- TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge (ICML 2024 · similarity 0.83)
- LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression (ICML 2025 · similarity 0.82)
- Auto-reconfiguration for Latency Minimization in CPU-based DNN Serving (ICML 2025 · similarity 0.82)
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.
