IVS News

Explore coverage by topic. Last 60 days.

All LPR VMS Edge Policy Lab Other

All Now On deck

Briefed only

92 items

Conditional Imaging: camera optical state per frame conceals the scene until authorized reconstruction

arxiv.org

October 22, 2026LabOn deckBriefed Oct 22

PRAISE: query-conditioned attribute subspaces lift VLM ReID at test time, no ID labels

arxiv.org

October 22, 2026LabOn deckBriefed Oct 22

arxiv.org

October 22, 2026LabOn deckBriefed Oct 22

RGB–IR fusion for counter-UAV: early fusion wins raw mAP; late fusion survives misaligned sensors

arxiv.org

October 22, 2026LabOn deckBriefed Oct 22

CodecMotion-OAD: decoder motion vectors replace optical flow for streaming action detection; 11.48× CPU

arxiv.org

October 21, 2026LabOn deckBriefed Oct 21

Foreground Attraction: weakly supervised camera-only 3D occupancy; foreground IoU 6.29 → 6.96

arxiv.org

October 21, 2026LabOn deckBriefed Oct 21

Jetson AGX Orin depth bake-off: NVIDIA ESS 0.70% AbsRel @ 8.6 ms; monocular models scatter 17–46% of edges

arxiv.org

October 21, 2026LabOn deckBriefed Oct 21

CGPLR: unsupervised cross-domain ReID; 76.4% Rank-1 Market→MSMT17, 95.1% MSMT17→Market

arxiv.org

October 20, 2026LabOn deckBriefed Oct 20

EMMA: event-driven multimodal memory for hours-long video; dual-stream boundaries, not fixed clips

arxiv.org

October 20, 2026LabOn deckBriefed Oct 20

GeoForward: sparse LiDAR anchors feedforward 3D Gaussian splatting; lowest nuScenes cross-camera depth error

arxiv.org

October 20, 2026LabOn deckBriefed Oct 20

YUTELA: Jetson Thor multi-site VAD; 31 CCTV cams / 170 h; TensorRT 14.17 → 29.94 FPS

arxiv.org

October 20, 2026LabOn deckBriefed Oct 20

ODSS-Net: orthogonal shared-structure / modality-specific ReID so alignment does not erase the extra camera

arxiv.org

October 19, 2026LabOn deckBriefed Oct 19

RCA-MOT: RGB–infrared MOT that does not crush both cameras into one vector; +5.4 / +6.1 HOTA

arxiv.org

October 19, 2026LabOn deckBriefed Oct 19

TriSAR-3D / PaintFusion: LiDAR + RGB + thermal 3D person detect; 65.4 AP@50, 51.6 ms; lying-pose models collapse without lying training

arxiv.org

October 19, 2026LabOn deckBriefed Oct 19

ARC-SORT: uncertainty-adaptive MOT; 81.82 HOTA SportsMOT, fewest ID switches of 69 entries

arxiv.org

October 16, 2026LabOn deckBriefed Oct 16

FALCON: KR260 FPGA occupancy with a bounded tail; Jetson Orin 4.4× median, L4 23.6×

arxiv.org

October 16, 2026LabOn deckBriefed Oct 16

MaLoW: streaming VLM that writes history into LoRA weights, not the context window

arxiv.org

October 16, 2026LabOn deckBriefed Oct 16

DeltaStreamer: agentic streaming video with change-aware memory + targeted replay

arxiv.org

October 15, 2026LabOn deckBriefed Oct 15

Detector architecture vs scene density: RT-DETR wins crowded MOT; YOLO26-L + NMS is the speed/quality split

arxiv.org

October 15, 2026LabOn deckBriefed Oct 15

LidarVerse: NVIDIA SIL real-time unified LiDAR world model; 11 FPS on H100

arxiv.org

October 15, 2026LabOn deckBriefed Oct 15

SAM3RL: RL memory controller for SAM 3 tracking; SOTA on 10 of 12 benches, no runtime tax

arxiv.org

October 15, 2026LabOn deckBriefed Oct 15

InfraWorld: roadside 4D occupancy as selective state evolution

arxiv.org

October 14, 2026LabOn deckBriefed Oct 14

SPARE: streaming VLM that gates history retrieval and shares the visual encoder

arxiv.org

October 14, 2026LabOn deckBriefed Oct 14

Temporal Evidence Selection: VAD explanations need object-tag–conditioned sampling, not anomaly saliency

arxiv.org

October 14, 2026LabOn deckBriefed Oct 14

When Trackers Fail: VLM Dual-Agent verifies SAM-class trackers and recovers identity drift

arxiv.org

October 14, 2026LabOn deckBriefed Oct 14

Event cameras for long action sequences: multi-frame MAE beats single-frame event stacks

arxiv.org

October 13, 2026LabOn deckBriefed Oct 13

MemReg: streaming outdoor LiDAR registration with hybrid pose memory (NeurIPS 2026)

arxiv.org

October 13, 2026LabOn deckBriefed Oct 13

SpatialASM: train LVLMs to revisit the scene from task-relevant viewpoints

arxiv.org

October 13, 2026LabOn deckBriefed Oct 13

Ultralytics 8.4.176: YOLO-World checkpoints 328 MB → 26 MB after set_classes

github.com

October 13, 2026LabBriefed Oct 13

AMGT: UAV maritime visible+IR+AIS vessel ID/track; MAir-YOLOv8-OBB + MAir-SORT

arxiv.org

October 12, 2026LabOn deckBriefed Oct 12

OccludeBench: VLMs miss what’s hidden; 42.7% on cue-based occlusion vs 62.7% inferential

arxiv.org

October 12, 2026LabOn deckBriefed Oct 12

UniMergeTrack: merge RGB-T / RGB-Event / RGB-Depth trackers instead of joint-training them

arxiv.org

October 12, 2026LabOn deckBriefed Oct 12

NXP & Forlinx: Empowering Industrial Edge AI with i.MX 95 and Ara240 M.2 AI Accelerator

timestech.in

October 10, 2026Lab

FastBench: streaming VLMs fail high-dynamic streams; Gemini-3.5-Flash 50.7%, 2 FPS misses the event

arxiv.org

October 9, 2026LabOn deckBriefed Oct 9

Onboard maritime YOLOX-S on Versal VC1902: tiles keep 98% of vessels at 29% of scene volume

arxiv.org

October 9, 2026LabOn deckBriefed Oct 9

Uni-DSM: one denoising score-matching backbone for VAD and VAA

arxiv.org

October 9, 2026LabBriefed Oct 9

VESSI: prompt-driven VLM interrogation of surveillance video; >66% flag rate, >85% review-time cut

arxiv.org

October 9, 2026LabBriefed Oct 9

BagDINO: DINOv3 + LoRA multi-view baggage ReID for mishandled airport bags

arxiv.org

October 8, 2026LabOn deckBriefed Oct 8

MOTIP2 (BMVC 2026): spatial priors fix E2E MOT ID switches; 73.4 HOTA DanceTrack

arxiv.org

October 8, 2026LabBriefed Oct 8

PanoPed: sim-to-real panoramic pedestrian tracking; Sextant head +2.19 HOTA on MOTIP

arxiv.org

October 8, 2026LabBriefed Oct 8

arXiv: CCDF — deepfake surveillance benchmark; SOTA detectors fail on Grok / VEO / Sora crime clips

arxiv.org

October 7, 2026LabOn deckBriefed Oct 7

arXiv: frozen VideoLLMs already encode evidence readiness for streaming QA

arxiv.org

October 7, 2026LabOn deckBriefed Oct 7

TwelveLabs Pegasus 1.6: egocentric video-to-text for physical AI, GA Oct 6

twelvelabs.io

October 7, 2026LabBriefed Oct 7

arXiv: RBA — explainable weakly-supervised video anomaly detection on MLLMs

arxiv.org

October 6, 2026LabOn deckBriefed Oct 6

arXiv: TRIM-ReID — RGB + NIR + thermal ReID with duplication-aware token drop

arxiv.org

October 6, 2026LabOn deckBriefed Oct 6

arXiv: referring MOT on a moving camera — residual motion after ego-motion subtract

arxiv.org

October 6, 2026LabOn deckBriefed Oct 6

arXiv: CHASE-VLA — post-training quantization for vision-language-action models

arxiv.org

October 5, 2026LabOn deckBriefed Oct 5

arXiv: FORESIGHT — training-free “plan the next look” for streaming VLMs

arxiv.org

October 5, 2026LabOn deckBriefed Oct 5

arXiv: Physical AI Smart Spaces — NVIDIA multi-camera 3D bench, ~1,800 cameras, 280+ hours

arxiv.org

October 5, 2026LabOn deckBriefed Oct 5

arXiv: ASH — zero-shot open-vocab MOT/segmentation on long video, <25 GB GPU

arxiv.org

October 2, 2026LabOn deckBriefed Oct 2

arXiv: FlashBack — training-free “when to remember” for streaming VLMs

arxiv.org

October 2, 2026LabOn deckBriefed Oct 2

arXiv: Video Evidence Indexing — cheap preview, then spend the high-res token budget on the moments that matter

arxiv.org

October 2, 2026LabOn deckBriefed Oct 2

arXiv: GaugeVLM — measured 3D interventions fix spatial VLM self-contradiction

arxiv.org

October 1, 2026LabOn deckBriefed Oct 1

arXiv: PhyProbe — physical-consistency scorer for generated video (NeurIPS 2026 poster)

arxiv.org

October 1, 2026LabOn deckBriefed Oct 1

arXiv: Video-RSI — video agents rewrite their own evidence harness

arxiv.org

September 30, 2026LabOn deckBriefed Sep 30

arXiv: VideoLoop — rewrite working memory so long-video agents stop thrashing

arxiv.org

September 30, 2026LabOn deckBriefed Sep 30

AMD buys Fei-Fei Li’s World Labs for $8.2 billion — spatial models inside the chip stack

ir.amd.com

September 29, 2026LabBriefed Sep 29

arXiv: EdgeVLN — 4-bit VLN on Jetson Orin NX 16 GB, 20.8× faster than streamed BF16

arxiv.org

September 29, 2026LabOn deckBriefed Sep 29

arXiv: Rift — Jetson AGX Orin VLM, −78% energy / −69% latency by dropping answer-invariant tiles

arxiv.org

September 28, 2026LabOn deckBriefed Sep 28

arXiv: TrafficFab — edge–cloud fabric for ~400 live CCTV streams (10% of Bangalore) on Jetson + fog

arxiv.org

September 28, 2026LabOn deckBriefed Sep 28

arXiv: CDKF-Track — cooperative 3D MOT for EdgeAI, up to +28% vs multi-agent SOTA

arxiv.org

September 25, 2026LabOn deckBriefed Sep 25

arXiv: GTR — softmax-free dense backbone, TensorRT on DRIVE AGX Thor in 2.3–8.8 ms

arxiv.org

September 25, 2026LabOn deckBriefed Sep 25

arXiv: LEAP-NBV — foundation-model next-best-view on a Jetson Xavier NX at 3.6 fps / 2.6 J

arxiv.org

September 23, 2026LabOn deckBriefed Sep 23

arXiv: FIVE-VLA — 641M driving VLA, ~4 fps on a T4 (edge proxy), 8–30× faster

arxiv.org

September 22, 2026LabOn deckBriefed Sep 22

arXiv: RoboDawn — agentic VLM drives a robot with translate / rotate / grip; one demo → SOTA

arxiv.org

September 22, 2026LabOn deckBriefed Sep 22

arXiv: TERMon — FPGA runtime monitor on edge AI, two-cycle decision, no BRAM/DSP

arxiv.org

September 22, 2026LabOn deckBriefed Sep 22

arXiv: PROVIA — online mistake detection on full egocentric trials, 58–70 fps

arxiv.org

September 18, 2026LabOn deckBriefed Sep 18

arXiv: Real-Time EXPO-FT — VLA proposes, a fast editor reacts; 42% → 97% in 10 minutes of robot data

arxiv.org

September 18, 2026LabOn deckBriefed Sep 18

arXiv: rMuscle — VLA “muscle memory” cache, 1.29–1.42× on Jetson Thor

arxiv.org

September 17, 2026LabOn deckBriefed Sep 17

arXiv: Tracking the Unseen — YOLOv11n + Kalman + ReID through full/long occlusion

arxiv.org

September 17, 2026LabOn deckBriefed Sep 17

arXiv: AVNet — one model classifies emergency vehicles from audio, video, or both

arxiv.org

September 16, 2026LabOn deckBriefed Sep 16

arXiv: PhysStream — streaming physics-grounded video with online scene memory + object tracks

arxiv.org

September 16, 2026LabOn deckBriefed Sep 16

arXiv: BVB / Blender-VideoBench — if the agent understood the video, it can rebuild it in Blender

arxiv.org

September 15, 2026LabOn deckBriefed Sep 15

arXiv: PhysBrain 1.5 — 8B VLM → physical foundation model (understand / act / predict)

arxiv.org

September 15, 2026LabOn deckBriefed Sep 15

arXiv: CFD — caption once on the edge, pull frames only when the question needs pixels

arxiv.org

September 14, 2026LabOn deckBriefed Sep 14

arXiv: EVPeriscope — ground robot’s event camera tracks a drone by its propellers

arxiv.org

September 14, 2026LabOn deckBriefed Sep 14

arXiv: FARM — 34k-param readout on a frozen robot world model flags failure

arxiv.org

September 14, 2026LabOn deckBriefed Sep 14

arXiv: MMGait — RGB + IR + depth + LiDAR + radar gait bench; one model, missing sensors

arxiv.org

September 11, 2026LabOn deckBriefed Sep 11

arXiv: Point4D — feed-forward 4D tracks across 200+ frames, not a 30-frame window

arxiv.org

September 11, 2026LabOn deckBriefed Sep 11

arXiv: LogiScope-VQA — warehouse VLM hazard bench; GPT-class models still miss human

arxiv.org

September 10, 2026LabOn deckBriefed Sep 10

arXiv: TempoGround — streaming VLM grounding (enter / stay / leave) then 3D box

arxiv.org

September 10, 2026LabOn deckBriefed Sep 10

arXiv: VLX-VR — agentic video reasoning with a Think–Memory–Observation loop

arxiv.org

September 10, 2026LabOn deckBriefed Sep 10

AGC-VLN: training-free UAV + UGV collab via a shared bird’s-eye map

arxiv.org

September 9, 2026LabOn deckBriefed Sep 9

Antioch raises $32M Series A (Greylock) — sim platform on NVIDIA Isaac, customer Ring

morningstar.com

September 9, 2026LabOn deckBriefed Sep 9

arXiv: MINT — world-space camera + hand motion from egocentric RGB (physical AI)

arxiv.org

September 9, 2026LabOn deckBriefed Sep 9

arXiv: ReactVAU — streaming anomaly understanding, heavy VLM only when the fast path fires

arxiv.org

September 9, 2026LabOn deckBriefed Sep 9

arXiv: VLM as a tracker tuner — no target labels, recovers most of the MOT17→MOT20 drop

arxiv.org

September 8, 2026LabOn deckBriefed Sep 8

arXiv: weakly supervised video anomaly detection at event level, not fixed top-k snippets

arxiv.org

September 8, 2026LabOn deckBriefed Sep 8

CSI-VAD training-free VLM VAD

arxiv.org

September 7, 2026LabOn deckBriefed Sep 7

InternVideo3 agentic long-video

arxiv.org

September 7, 2026LabOn deckBriefed Sep 7

Scientific Reports: YOLOv8n hits 96.3% mAP on an 8-class in-cabin driver-activity set

nature.com

September 1, 2026LabBriefed Sep 1

YOLO26: NMS-free, 7-task family, ~43% faster on CPU — and AGPL still bites IVS products

docs.ultralytics.com

August 26, 2026LabBriefed Aug 26