Papers

3

Total Citations

11

H-Index

2

About

Jianing Qian is a researcher at the forefront of efficient deep learning and robot perception, with a focus on making computer vision models both faster and more practical for real-world deployment. Her work bridges the gap between high-performance neural networks and resource-constrained systems, particularly in robotics. Qian’s major contributions include pioneering the **Depth-wise Decomposition** technique for accelerating separable convolutions in CNNs, a method that reduces inference latency without sacrificing accuracy—critical for real-time applications. She also advanced the use of generic vision transformers as object-centric scene encoders for manipulation policies, showing how pre-trained models can be repurposed for robotic tasks without specialized training. Additionally, her work on **FusionMapping** combines monocular images with 2D laser scans to predict depth, offering a cost-effective alternative to expensive LiDAR. With over 11 citations across her key papers, Qian’s research is gaining traction for its practical impact on efficient vision systems. Her 2023 paper on depth-wise decomposition and her 2024 work on transformer-based scene encoding highlight her ability to innovate at the intersection of computer vision and robotics, making her a rising voice in efficient AI.

Research Focus

Key Achievements

2
H-Index
3
Papers
11
Total Citations
4
Avg Citations/Paper
🏆 Most Cited Paper
Depth-wise Decomposition for Accelerating Separable Convolutions in Efficient Convolutional Neural Networks
5 citations · 2023
📈 Most Prolific Year: 2023 (1 Papers)
🤝 Key Collaborators: 7
🏛 Institutions: Carnegie Mellon University, University of Pennsylvania

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago