-
GPT
Binhua Huang is driven by a simple conviction: true intelligence isn't about using more resources, but achieving more with less. Ambitious in vision, relentless in execution, and unwilling to settle for ordinary answers.
-
Claude
Where others decode every frame, Binhua reads the motion vectors. He finds the story in what the codec already knew.
-
Gemini
Binhua Huang bridges the gap between artificial intelligence and the physical world, giving machines the gift of true perception.
-
Grok
Binhua Huang is the kind of researcher who quietly makes visual AI leaner and more practical, one motion vector at a time.
๐ฅ News
- 2026.10: ย ๐ MoCLIP has been accepted at The Visual Computer! [paper] [code]
- 2026.09: ย ๐ Optimal Brain Connection, has been accepted at IEEE Transactions on Multimedia and is now available on: [paper]
- 2026.07: ย ๐ EarthCamQA, has been accepted at iEMSs and is now available on: [paper] [code]
- 2026.03: ย ๐ MoCrop, has been accepted at IEEE IJCNN. [paper] [code]
- 2026.02: ย ๐ MoCLIP-Lite, has been accepted at IEEE CISP and is now available on IEEE Xplore: [paper] [code]
- 2026.01: ย ๐ TinyDrop has been accepted at IEEE ICASSP! [paper]
- 2025.10: ย ๐ค Contributed MV-Only (high-speed mode) to mv-extractor. [upstream] [my fork] [PyPI]
- 2025.09: ย ๐ Preprinted MVP on arXiv: [paper] [code]
๐ Latest Publications
2026
-
Training-Efficient Video Action Recognition Through Vision-Language Semantic Priors and Compressed-Domain Motion Fusion
B. Huang, Z. Cui, W. Yao, Y. Li, and S. Dev
The Visual Computer (accepted) ยท [paper] [code] -
Optimal brain connection: Towards efficient structural pruning
S. Chen, W. Ma, B. Huang, Q. Wang, G. Wang, W. Sun, L. Huang, D. John
IEEE Transactions on Multimedia [paper] [arXiv] [code] -
EarthCamQA: An Agentic LLMโVLM Framework for Real-Time Visual Question Answering in Urban Environments
B. Huang, W. Yao, Y. Li, D. A. Cabrera, S. Dev
iEMSs 2026 [paper] [code] -
MoCrop: Training Free Motion Guided Cropping for Efficient Video Action Recognition
B. Huang, W. Yao, S. Chen, G. Wang, Q. Wang, S. Dev
IEEE IJCNN 2026 [paper] [code] -
TinyDrop: Tiny Model Guided Token Dropping for Vision Transformers
G. Wang, Q. Wang, B. Huang, S. Chen, D. John
IEEE ICASSP 2026 [paper] [arXiv]
2025
-
MoCLIP-Lite: Efficient Video Recognition by Fusing CLIP with Motion Vectors
B. Huang, N. Wang, A. Pakrashi, S. Dev
IEEE CISP 2025 [paper] [code] -
MVP: Motion Vector Propagation for Zero-Shot Video Object Detection
B. Huang, N. Wang, W. Yao, S. Dev
arXiv preprint [paper] [code] -
DCentNet: Decentralized multistage biomedical signal classification using early exits
X. Li, B. Huang, B. Cardiff, D. John
Biomedical Signal Processing and Control [paper] [code]
2024
- Dynamic liquid volume estimation using optical tactile sensors and spiking neural network
B. Huang, S. Fang, M. Yin et al.
Springer - Intelligent Service Robotics [paper]
๐งฉ Open-Source Projects & Contributions
-
mv-extractor โ Contributor (2025)
A fast H.264/MPEG-4 motion-vector extractor with optional RGB decode skipping.
โข Key: Implemented Motion-Vectors-Only / frame-decode skipping mode; docs & tests; acknowledged upstream.
โข Links: Upstream ยท My fork ยท PyPI
-
GelPixel โ Tactile Sensing (RCAR 2023)
A single-pixel tactile sensor research release, including the dataset, multi-target regression code, Fourier parameters, and sensor-structure comparison notebook.
โข Links: Repository ยท Paper ยท Dataset
-
CNC_5axis โ Motion Control
A Python interface for five-axis CNC control over serial communication, supporting distance-based movement, timed movement, direction selection, and speed adjustment.
โข Links: Repository
-
Cloud IoT System โ Embedded IoT
An IoT prototype connecting 8051/STM32 devices through ESP8266 WiFi to TCP cloud services for temperature monitoring, data logging, and remote control; includes firmware and hardware diagrams.
โข Links: Repository ยท Hardware diagrams
๐จโ๐ป Service
- Session Chair: IJCNN2026
- Program Committee: AAAI2027, AAAI2026, SWC2026
- Session Assistant: RCAR2023
- Reviewer: IEEE T-PAMI, IEEE T-IP, IEEE T-MM, IEEE T-ASE, IEEE T-IM, AAAI, ACM MM, ACM WWW, IEEE ICRA, IEEE ICASSP, IEEE IJCNN, BMVC
๐จโ๐ป Research Focuses
โข Video Understanding (Outputs) 2025 - present โข Model Compression (Outputs) 2023 - 2025 โข Embodied Intelligence (Outputs) 2021 - 2023
๐โโ๏ธ Personal life
- ๐โโ๏ธ Completed two marathons (42.195 km)