Post
2403
Don't sleep on new AI at Meta Vision-Language release! ๐ฅ
facebook/perception-encoder-67f977c9a65ca5895a7f6ba1
facebook/perception-lm-67f9783f171948c383ee7498
Meta dropped swiss army knives for vision with A2.0 license ๐
> image/video encoders for vision language modelling and spatial understanding (object detection etc) ๐
> The vision LM outperforms InternVL3 and Qwen2.5VL ๐
> They also release gigantic video and image datasets
The authors attempt to come up with single versatile vision encoder to align on diverse set of tasks.
They trained Perception Encoder (PE) Core: a new state-of-the-art family of vision encoders that can be aligned for both vision-language and spatial tasks. For zero-shot image tasks, it outperforms latest sota SigLIP2 ๐
> Among fine-tuned ones, first one is PE-Spatial. It's a model to detect bounding boxes, segmentation, depth estimation and it outperforms all other models ๐ฎ
> Second one is PLM, Perception Language Model, where they combine PE-Core with Qwen2.5 LM 7B. it outperforms all other models (including InternVL3 which was trained with Qwen2.5LM too!)
The authors release the following checkpoints in sizes base, large and giant:
> 3 PE-Core checkpoints (224, 336, 448)
> 2 PE-Lang checkpoints (L, G)
> One PE-Spatial (G, 448)
> 3 PLM (1B, 3B, 8B)
> Datasets
Authors release following datasets ๐
> PE Video: Gigantic video datasete of 1M videos with 120k expert annotations โฏ๏ธ
> PLM-Video and PLM-Image: Human and auto-annotated image and video datasets on region-based tasks
> PLM-VideoBench: New video benchmark on MCQA
facebook/perception-encoder-67f977c9a65ca5895a7f6ba1
facebook/perception-lm-67f9783f171948c383ee7498
Meta dropped swiss army knives for vision with A2.0 license ๐
> image/video encoders for vision language modelling and spatial understanding (object detection etc) ๐
> The vision LM outperforms InternVL3 and Qwen2.5VL ๐
> They also release gigantic video and image datasets
The authors attempt to come up with single versatile vision encoder to align on diverse set of tasks.
They trained Perception Encoder (PE) Core: a new state-of-the-art family of vision encoders that can be aligned for both vision-language and spatial tasks. For zero-shot image tasks, it outperforms latest sota SigLIP2 ๐
> Among fine-tuned ones, first one is PE-Spatial. It's a model to detect bounding boxes, segmentation, depth estimation and it outperforms all other models ๐ฎ
> Second one is PLM, Perception Language Model, where they combine PE-Core with Qwen2.5 LM 7B. it outperforms all other models (including InternVL3 which was trained with Qwen2.5LM too!)
The authors release the following checkpoints in sizes base, large and giant:
> 3 PE-Core checkpoints (224, 336, 448)
> 2 PE-Lang checkpoints (L, G)
> One PE-Spatial (G, 448)
> 3 PLM (1B, 3B, 8B)
> Datasets
Authors release following datasets ๐
> PE Video: Gigantic video datasete of 1M videos with 120k expert annotations โฏ๏ธ
> PLM-Video and PLM-Image: Human and auto-annotated image and video datasets on region-based tasks
> PLM-VideoBench: New video benchmark on MCQA
