Camera-Only AI Sees Past Occlusions in Real Time for Self-Driving Cars
Researchers in Chongqing have built a lightweight camera-only 3D detection framework that adapts its processing to scene complexity, fuses information ...
Researchers in Chongqing have built a lightweight camera-only 3D detection framework that adapts its processing to scene complexity, fuses information ...
Researchers have developed AttentiveGaze, a real-time multimodal AI framework that fuses face, eye, and head pose cues with attention mechanisms ...
Researchers at IIITDM Kancheepuram have developed DECA-YOLO, a lightweight dual-pooling channel attention enhancement to YOLOv7 that improves real-time pedestrian detection ...
Researchers at the University of Leeds have developed PICO, a single-stage AI system that estimates the full 3D position and ...
A new stem-encoder-decoder neural network runs crowd counting at up to 381.7 frames per second on desktop GPUs and 71.9 ...
A new deep learning framework called EEG-SDL-Net combines Kolmogorov–Arnold networks, Sobel-based temporal enhancement, and channel attention to classify EEG signals ...
Researchers have developed LightViT-AD, a compact vision transformer framework that detects anomalies in aerial imagery in real time on drone ...
A new Nature study presents Delphy, a GPU-accelerated Bayesian phylogenetics platform that delivers dated, uncertainty-aware viral trees from outbreak-scale datasets ...
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine