Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024)
-
Updated
Apr 9, 2025 - Jupyter Notebook
#
ego4d
Here are 10 public repositories matching this topic...
NaQ: Leveraging Narrations as Queries to Supervise Episodic Memory. CVPR 2023.
-
Updated
Jan 26, 2024 - Python
Models of Mental Simulation
computer-vision transformers neuroscience recurrent-neural-networks computational-neuroscience vision convolutional-neural-networks computational-cognitive-science pytorch-implementation ego4d
-
Updated
Jul 25, 2024 - Jupyter Notebook
Winner of CVPR23 EGO4D STA challenge
-
Updated
Oct 14, 2024 - Python
Curated datasets, benchmarks, models, and tools for egocentric AI, embodied intelligence, VLA, world models, robotics, and wearable vision.
research computer-vision robotics augmented-reality benchmarks awesome-list datasets video-understanding robot-learning world-models embodied-ai hand-object-interaction video-language egocentric-vision first-person-vision ego4d vision-language-action multimodal-ai wearable-camera egocentric-ai
-
Updated
Aug 23, 2026 - Ruby
[NeurIPS 2026] Linguistic Trajectory Encoding (LTE): object-centric spatiotemporal memory for embodied agents. CPU and real-video demos with SAM3, ViPE, DINOv2, and Qwen3-VL. Spatial Memory Benchmark (SMB). arXiv:2609.04802.
video-understanding long-term-memory embodied-ai ego4d spatiotemporal-memory trajectory-compression neurips-2026 linguistic-trajectory-encoding object-centric-memory spatial-memory-benchmark egolife natural-language-retrieval
-
Updated
Sep 29, 2026 - Python
A two-step framework for extracting textual answers from egocentric videos via NLQ, combining VSLNet for segment localization and Video-LLaVA for efficient answer generation.
nlq episodic-memory video-retrieval egocentric-videos ego4d vision-language-models natural-language-querying video-llava
-
Updated
Mar 9, 2026 - Python
Official implementation of FlowNar: Scalable Streaming Narration for Long-Form Videos
-
Updated
Jun 29, 2026 - Python
Efficient NLVL on Ego4D. Benchmark VSLBase/VSLNet with BERT/GloVE, EgoVLP/Omnivore, configurable FiLM conditioning layer (Perez et al., FiLM: Visual Reasoning with a General Conditioning Layer), explores compression via KD, including CBKD (Lan et al., Counterclockwise Block‑by‑Block Knowledge Distillation), prototype post‑training quantization.
-
Updated
Jul 19, 2026 - Jupyter Notebook
Two-stage pipeline on the Ego4D NLQ benchmark: finding the moment in a first-person video that answers a question, then answering it — VSLNet/VSLBase ablation for temporal localisation, Video-LLaVA for answer generation, scored against hand-written ground truth.
deep-learning pytorch episodic-memory video-question-answering ego4d temporal-localization video-llava multimodal-ai vslnet
-
Updated
Sep 16, 2026 - Jupyter Notebook
Add this topic to your repo
To associate your repository with the ego4d topic, visit your repo's landing page and select "manage topics."