Back to papers
Jointly Optimizing Preprocessing and Inference for DNN-based Visual Analytics
Summary: End-to-end DNN for visual analytics; preprocessing bottlenecks. SMOL: using native low-res data and a runtime that pipelines preprocessing and inference with CPU/GPU placement, memory threading, delivering up to 5.9× throughput at fixed accuracy.
(summarized by gpt-5-nano on Feb 09 2026)
- Paper ID
- 12603
- Venue
- VLDB
- Year
- 2021
- Pagerank
- 7.2607192e-05
- Overall Rank
- 3,291 | 77.13%
- DOI
-
10.14778/3425879.3425881
Incoming Non-self Citations Over Time
Incoming Citations (Sorted by Pagerank)
Showing 14 of 14 citing papers.
| Rank |
Citing Paper |
Year |
Venue |
Pagerank |
| 2,175 |
tf.data: A Machine Learning Data Processing Framework |
2021 |
VLDB |
9.3745231e-05 |
| 3,409 |
End-to-end Optimization of Machine Learning Prediction Queries |
2022 |
SIGMOD |
7.1240791e-05 |
| 3,694 |
Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines |
2022 |
SIGMOD |
6.8316905e-05 |
| 4,265 |
VSS: A Storage System for Video Analytics |
2021 |
SIGMOD |
6.3012892e-05 |
| 4,492 |
TASTI: Semantic Indexes for Machine Learning-based Queries over Unstructured Data |
2022 |
SIGMOD |
6.1374891e-05 |
| 4,565 |
Optimizing Video Analytics with Declarative Model Relationships |
2023 |
VLDB |
6.0746821e-05 |
| 4,642 |
VIVA: An End-to-End System for Interactive Video Analytics |
2022 |
CIDR |
6.0214283e-05 |
| 4,703 |
Accelerating Approximate Aggregation Queries with Expensive Predicates |
2021 |
VLDB |
5.9793615e-05 |
| 7,911 |
Accelerating Aggregation Queries on Unstructured Streams of Data |
2023 |
VLDB |
4.6143141e-05 |
| 8,646 |
Optimizing Video Selection LIMIT Queries With Commonsense Knowledge |
2024 |
VLDB |
4.4720866e-05 |
| 9,677 |
Apt-Serve: Adaptive Request Scheduling on Hybrid Cache for Scalable LLM Inference Serving |
2025 |
SIGMOD |
4.3006524e-05 |
| 9,767 |
TVM: A Tile-based Video Management Framework |
2024 |
VLDB |
4.2815042e-05 |
| 9,787 |
The Image Calculator: 10x Faster Image-AI Inference by Replacing JPEG with Self-designing Storage Format |
2024 |
SIGMOD |
4.2799988e-05 |
| 10,947 |
Predictive and Near-Optimal Sampling for View Materialization in Video Databases |
2024 |
SIGMOD |
4.1905499e-05 |
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Semantically Similar Papers
| Overall Rank |
Paper |
Year |
Venue |
Pagerank |
| 4,175 |
FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline |
2023 |
VLDB |
6.3772575e-05 |
| 9,619 |
Faster Visual Analytics through Pixel-Perfect Aggregation |
2014 |
VLDB |
4.3129474e-05 |
| 9,704 |
ETO: Accelerating Optimization of DNN Operators by High-Performance Tensor Program Reuse |
2022 |
VLDB |
4.2953234e-05 |
| 13,242 |
Using Deep Learning Models to Replace Large Materialized Views in Relational Database |
2021 |
CIDR |
- |
| 2,892 |
VISTA: Optimized System for Declarative Feature Transfer from Deep CNNs at Scale |
2020 |
SIGMOD |
7.9570135e-05 |
| 1,655 |
Challenges and Opportunities in DNN-Based Video Analytics: A Demonstration of the BlazeIt Video Query Engine |
2019 |
CIDR |
0.000109771 |
| 2,871 |
Incremental and Approximate Inference for Faster Occlusion-based Deep CNN Explanations |
2019 |
SIGMOD |
7.9800983e-05 |
| 2,536 |
DeepLens: Towards a Visual Data Management System |
2019 |
CIDR |
8.5817195e-05 |
| 1,500 |
Analyzing and Mitigating Data Stalls in DNN Training |
2021 |
VLDB |
0.00011636174 |
| 3,694 |
Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines |
2022 |
SIGMOD |
6.8316905e-05 |